Log In Sign Up
Models:
Joel Williams
JoelW/
pymskt
0
Downloads: 1
Diff of /docs/image/cartilage_processing.html [000000] .. [9173ee]

Switch to unified view

a b/docs/image/cartilage_processing.html 

  1
  
<!doctype html>





  
  

  2
  
<html lang="en">





  
  

  3
  
<head>





  
  

  4
  
<meta charset="utf-8">





  
  

  5
  
<meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1" />





  
  

  6
  
<meta name="generator" content="pdoc 0.10.0" />





  
  

  7
  
<title>pymskt.image.cartilage_processing API documentation</title>





  
  

  8
  
<meta name="description" content="" />





  
  

  9
  
<link rel="preload stylesheet" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/11.0.1/sanitize.min.css" integrity="sha256-PK9q560IAAa6WVRRh76LtCaI8pjTJ2z11v0miyNNjrs=" crossorigin>





  
  

  10
  
<link rel="preload stylesheet" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/11.0.1/typography.min.css" integrity="sha256-7l/o7C8jubJiy74VsKTidCy1yBkRtiUGbVkYBylBqUg=" crossorigin>





  
  

  11
  
<link rel="stylesheet preload" as="style" href="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/10.1.1/styles/github.min.css" crossorigin>





  
  

  12
  
<style>:root{--highlight-color:#fe9}.flex{display:flex !important}body{line-height:1.5em}#content{padding:20px}#sidebar{padding:30px;overflow:hidden}#sidebar > *:last-child{margin-bottom:2cm}.http-server-breadcrumbs{font-size:130%;margin:0 0 15px 0}#footer{font-size:.75em;padding:5px 30px;border-top:1px solid #ddd;text-align:right}#footer p{margin:0 0 0 1em;display:inline-block}#footer p:last-child{margin-right:30px}h1,h2,h3,h4,h5{font-weight:300}h1{font-size:2.5em;line-height:1.1em}h2{font-size:1.75em;margin:1em 0 .50em 0}h3{font-size:1.4em;margin:25px 0 10px 0}h4{margin:0;font-size:105%}h1:target,h2:target,h3:target,h4:target,h5:target,h6:target{background:var(--highlight-color);padding:.2em 0}a{color:#058;text-decoration:none;transition:color .3s ease-in-out}a:hover{color:#e82}.title code{font-weight:bold}h2[id^="header-"]{margin-top:2em}.ident{color:#900}pre code{background:#f8f8f8;font-size:.8em;line-height:1.4em}code{background:#f2f2f1;padding:1px 4px;overflow-wrap:break-word}h1 code{background:transparent}pre{background:#f8f8f8;border:0;border-top:1px solid #ccc;border-bottom:1px solid #ccc;margin:1em 0;padding:1ex}#http-server-module-list{display:flex;flex-flow:column}#http-server-module-list div{display:flex}#http-server-module-list dt{min-width:10%}#http-server-module-list p{margin-top:0}.toc ul,#index{list-style-type:none;margin:0;padding:0}#index code{background:transparent}#index h3{border-bottom:1px solid #ddd}#index ul{padding:0}#index h4{margin-top:.6em;font-weight:bold}@media (min-width:200ex){#index .two-column{column-count:2}}@media (min-width:300ex){#index .two-column{column-count:3}}dl{margin-bottom:2em}dl dl:last-child{margin-bottom:4em}dd{margin:0 0 1em 3em}#header-classes + dl > dd{margin-bottom:3em}dd dd{margin-left:2em}dd p{margin:10px 0}.name{background:#eee;font-weight:bold;font-size:.85em;padding:5px 10px;display:inline-block;min-width:40%}.name:hover{background:#e0e0e0}dt:target .name{background:var(--highlight-color)}.name > span:first-child{white-space:nowrap}.name.class > span:nth-child(2){margin-left:.4em}.inherited{color:#999;border-left:5px solid #eee;padding-left:1em}.inheritance em{font-style:normal;font-weight:bold}.desc h2{font-weight:400;font-size:1.25em}.desc h3{font-size:1em}.desc dt code{background:inherit}.source summary,.git-link-div{color:#666;text-align:right;font-weight:400;font-size:.8em;text-transform:uppercase}.source summary > *{white-space:nowrap;cursor:pointer}.git-link{color:inherit;margin-left:1em}.source pre{max-height:500px;overflow:auto;margin:0}.source pre code{font-size:12px;overflow:visible}.hlist{list-style:none}.hlist li{display:inline}.hlist li:after{content:',\2002'}.hlist li:last-child:after{content:none}.hlist .hlist{display:inline;padding-left:1em}img{max-width:100%}td{padding:0 .5em}.admonition{padding:.1em .5em;margin-bottom:1em}.admonition-title{font-weight:bold}.admonition.note,.admonition.info,.admonition.important{background:#aef}.admonition.todo,.admonition.versionadded,.admonition.tip,.admonition.hint{background:#dfd}.admonition.warning,.admonition.versionchanged,.admonition.deprecated{background:#fd4}.admonition.error,.admonition.danger,.admonition.caution{background:lightpink}</style>





  
  

  13
  
<style media="screen and (min-width: 700px)">@media screen and (min-width:700px){#sidebar{width:30%;height:100vh;overflow:auto;position:sticky;top:0}#content{width:70%;max-width:100ch;padding:3em 4em;border-left:1px solid #ddd}pre code{font-size:1em}.item .name{font-size:1em}main{display:flex;flex-direction:row-reverse;justify-content:flex-end}.toc ul ul,#index ul{padding-left:1.5em}.toc > ul > li{margin-top:.5em}}</style>





  
  

  14
  
<style media="print">@media print{#sidebar h1{page-break-before:always}.source{display:none}}@media print{*{background:transparent !important;color:#000 !important;box-shadow:none !important;text-shadow:none !important}a[href]:after{content:" (" attr(href) ")";font-size:90%}a[href][title]:after{content:none}abbr[title]:after{content:" (" attr(title) ")"}.ir a:after,a[href^="javascript:"]:after,a[href^="#"]:after{content:""}pre,blockquote{border:1px solid #999;page-break-inside:avoid}thead{display:table-header-group}tr,img{page-break-inside:avoid}img{max-width:100% !important}@page{margin:0.5cm}p,h2,h3{orphans:3;widows:3}h1,h2,h3,h4,h5,h6{page-break-after:avoid}}</style>





  
  

  15
  
<script defer src="https://cdnjs.cloudflare.com/ajax/libs/highlight.js/10.1.1/highlight.min.js" integrity="sha256-Uv3H6lx7dJmRfRvH8TH6kJD1TSK1aFcwgx+mdg3epi8=" crossorigin></script>





  
  

  16
  
<script>window.addEventListener('DOMContentLoaded', () => hljs.initHighlighting())</script>





  
  

  17
  
</head>





  
  

  18
  
<body>





  
  

  19
  
<main>





  
  

  20
  
<article id="content">





  
  

  21
  
<header>





  
  

  22
  
<h1 class="title">Module <code>pymskt.image.cartilage_processing</code></h1>





  
  

  23
  
</header>





  
  

  24
  
<section id="section-intro">





  
  

  25
  
<details class="source">





  
  

  26
  
<summary>





  
  

  27
  
<span>Expand source code</span>





  
  

  28
  
</summary>





  
  

  29
  
<pre><code class="python">import SimpleITK as sitk





  
  

  30
  
import numpy as np





  
  

  31
  
from scipy import ndimage as ndi





  
  

  32
  





  
  

  33
  





  
  

  34
  
def CofM(array):





  
  

  35
  
    &#39;&#39;&#39;





  
  

  36
  
    Get center of mass for a row of a binary 2D image. 





  
  

  37
  
    Parameters





  
  

  38
  
    ----------





  
  

  39
  
    array : 1D array





  
  

  40
  
        Individual row of a 2D image.  





  
  

  41
  
    Returns





  
  

  42
  
    -------





  
  

  43
  
    centerPixels : 





  
  

  44
  
        Average location of 1s in the row





  
  

  45
  
    Notes





  
  

  46
  
    -----





  
  

  47
  
    Calculates the average location of cartilage for the row of image being analyzed. 





  
  

  48
  
    Returns 0 if there are no pixels





  
  

  49
  
    





  
  

  50
  
    &#39;&#39;&#39;





  
  

  51
  
    pixels = np.where(array==1)





  
  

  52
  
    centerPixels = np.mean(pixels)





  
  

  53
  
    nans = np.isnan(centerPixels)





  
  

  54
  
    if nans == True:





  
  

  55
  
        centerPixels = 0





  
  

  56
  
    return(centerPixels)





  
  

  57
  





  
  

  58
  
def get_y_CofM(flattenedSeg):





  
  

  59
  
    &#39;&#39;&#39;





  
  

  60
  
    Get CofM of femoral cartilage for each row of the flattened segmentation. 





  
  

  61
  
    Parameters





  
  

  62
  
    ----------





  
  

  63
  
    flattenedSeg : 2D array





  
  

  64
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  65
  
    Returns





  
  

  66
  
    -------





  
  

  67
  
    yCofM : 





  
  

  68
  
        Find the CofM for each row of the image. 





  
  

  69
  
    Notes





  
  

  70
  
    -----





  
  

  71
  
    Get the x/y coordinates for the CofM for each row of the flattened segmentation. 





  
  

  72
  
    





  
  

  73
  
    &#39;&#39;&#39;





  
  

  74
  
    locationFemur = np.where(flattenedSeg==1)





  
  

  75
  
    yCofM = np.zeros((flattenedSeg.shape[0], 2), dtype=int)





  
  

  76
  
    





  
  

  77
  
    # only calculate for rows with cartilage. 





  
  

  78
  
    minRow = np.min(locationFemur[0])





  
  

  79
  
    maxRow = np.max(locationFemur[0])





  
  

  80
  
    





  
  

  81
  
    # iterate over rows of image, get CofM, store CofM for row. 





  
  

  82
  
    for x in range(minRow, maxRow):





  
  

  83
  
        yCofM[x, 0] = x #store the x-coordinate (row) we calcualted CofM for. 





  
  

  84
  
        yCofM[x, 1] = int(CofM(flattenedSeg[x, :])) # store the CofM value (make it an integer for indexing)





  
  

  85
  
    yCofM = yCofM[minRow+10:maxRow-10,:] # remove 10 most medial and most lateral pixels of femoral cartilage. 





  
  

  86
  
    return(yCofM) 





  
  

  87
  





  
  

  88
  
def absolute_CofM(flattenedSeg):





  
  

  89
  
    &#39;&#39;&#39;





  
  

  90
  
    Get absolute CofM of all the femoral cartilage pixels





  
  

  91
  
    Parameters





  
  

  92
  
    ----------





  
  

  93
  
    flattenedSeg : 2D array





  
  

  94
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  95
  
    Returns





  
  

  96
  
    -------





  
  

  97
  
    centerX : 





  
  

  98
  
        The CofM in the X direction for the segmentation  





  
  

  99
  
    centerY : 





  
  

  100
  
        The CofM in the Y direction for the segmentation





  
  

  101
  
    Notes





  
  

  102
  
    -----





  
  

  103
  
    Get the x/y coordinates for the CofM for the whole flattened segmentation





  
  

  104
  
    





  
  

  105
  
    &#39;&#39;&#39;





  
  

  106
  
    femurPoints = np.where(flattenedSeg==1)





  
  

  107
  
    centerX = np.mean(femurPoints[0])





  
  

  108
  
    centerY = np.mean(femurPoints[1])





  
  

  109
  
    return(centerX, centerY)





  
  

  110
  





  
  

  111
  
def findNotch(flattenedSeg, trochleaPositionX=1000):





  
  

  112
  
    &#39;&#39;&#39;





  
  

  113
  
    Get the X Y position of the trochlear notch - where medial/lateral sides of the femur meet. 





  
  

  114
  
    Parameters





  
  

  115
  
    ----------





  
  

  116
  
    flattenedSeg : 2D array





  
  

  117
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  118
  
    Returns





  
  

  119
  
    -------





  
  

  120
  
    trochleaPositionY : 





  
  

  121
  
        Y position of trochlear notch  





  
  

  122
  
    trochleaPositionX : 





  
  

  123
  
        X position of trochlear notch





  
  

  124
  
    Notes





  
  

  125
  
    -----





  
  

  126
  
    Get the x/y coordinates for the trochlear notch. This is an iterative method that assumes things about the shape the





  
  

  127
  
    femoral cartilage.





  
  

  128
  
    





  
  

  129
  
    &#39;&#39;&#39;





  
  

  130
  
    # Goal is to find the most anterior point that is between the medial/lateral condyles





  
  

  131
  





  
  

  132
  
    # First guess at the troch notch in the 1st axis (med/lat axis) is the location with the smallest value for





  
  

  133
  
    # the 2nd axis CofM. This is because in axis 1, negative is anterior and we expect the most anterior CofM should 





  
  

  134
  
    # roughly align with the trochlear notch.





  
  

  135
  
    y_CofM = get_y_CofM(flattenedSeg)





  
  

  136
  
    first_guess = y_CofM[np.argmin(y_CofM[:,1]), 0]





  
  

  137
  
    # the second guess is just the CofM of the whole cartilage. 





  
  

  138
  
    centerX, centerY = absolute_CofM(flattenedSeg)





  
  

  139
  
    second_guess = centerX





  
  

  140
  





  
  

  141
  
    # We use the 2 guesses to help define a search space for the trochlear notch.





  
  

  142
  
    min_search = int(np.min((first_guess,second_guess))-20)





  
  

  143
  
    max_search = int(np.max((first_guess,second_guess))+20)





  
  

  144
  





  
  

  145
  
    # now, we iterate over all of the rows (axis 1) of the search space (moving in the medial/lateral direction)





  
  

  146
  
    # we are looking for the row where the most posterior point (back of femur) is furthest anterior (notch). 





  
  

  147
  
    for y in range(min_search, max_search):





  
  

  148
  
        # At each row, we find most posterior pixel labeled as cartilage. 





  
  

  149
  
        try:





  
  

  150
  
            trochleaPosition_test = np.max(np.where(flattenedSeg[y,:]==1))





  
  

  151
  
        except ValueError:





  
  

  152
  
            # if there is no cartilage we&#39;ll get a ValueError exception. 





  
  

  153
  
            # in that case, set this value to be the max it can be (the size of the first axis)





  
  

  154
  
            trochleaPosition_test = flattenedSeg.shape[1]





  
  

  155
  
        # if the most posterior point for this row is more anterior than the current trochleaPositionX,





  
  

  156
  
        # then update this to be the new trochlear notch.





  
  

  157
  
        if trochleaPosition_test &lt; trochleaPositionX:





  
  

  158
  
            trochleaPositionX = trochleaPosition_test





  
  

  159
  
            trochleaPositionY = y





  
  

  160
  





  
  

  161
  
    return(trochleaPositionY, trochleaPositionX+1)





  
  

  162
  





  
  

  163
  
def getAnteriorOfWeightBearing(segArray, femurIndex=1):





  
  

  164
  
    &#39;&#39;&#39;





  
  

  165
  
    Prepare full segmentation and extract the trochlear notch location. 





  
  

  166
  
    Parameters





  
  

  167
  
    ----------





  
  

  168
  
    flattenedSeg : 2D array





  
  

  169
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  170
  
    femurIndex : int





  
  

  171
  
        Index of the label used to localize the femur in the array. 





  
  

  172
  
    Returns





  
  

  173
  
    -------





  
  

  174
  
    trochleaPositionY : 





  
  

  175
  
        Y position of trochlear notch  





  
  

  176
  
    trochleaPositionX : 





  
  

  177
  
        X position of trochlear notch





  
  

  178
  
    Notes





  
  

  179
  
    -----





  
  

  180
  
    Get the x/y coordinates for the trochlear notch. This is an iterative method that assumes things about the shape the femoral cartilage. 





  
  

  181
  
    First flatten and fill any holes in the segmentation. 





  
  

  182
  
    





  
  

  183
  
    &#39;&#39;&#39;





  
  

  184
  





  
  

  185
  
    femurSegmentation = np.zeros_like(segArray)





  
  

  186
  
    femurSegmentation[segArray == femurIndex] = 1





  
  

  187
  
    flattenedSegmentation = np.amax(femurSegmentation, axis=1)





  
  

  188
  
    flattened_seg_filled = ndi.binary_fill_holes(flattenedSegmentation)





  
  

  189
  
    trochY, trochX = findNotch(flattened_seg_filled)





  
  

  190
  
    return(trochY, trochX)





  
  

  191
  
    





  
  

  192
  
def getCartilageSubRegions(segArray, anteriorWBslice, posteriorWBslice, trochY,





  
  

  193
  
                           femurLabel=1, medTibiaLabel=2, latTibiaLabel=3, antFemurMask=5, 





  
  

  194
  
                           medWbFemurMask=6, latWbFemurMask=7, medPostFemurMask=8, latPostFemurMask=9):





  
  

  195
  
    &#39;&#39;&#39;





  
  

  196
  
    Take cartilage segmentation, and decompose femoral cartilage into subregions of interest.  





  
  

  197
  
    Parameters





  
  

  198
  
    ----------





  
  

  199
  
    segArray : array





  
  

  200
  
        3D array with segmentation for the cartialge regions. 





  
  

  201
  
    anteriorWBslice : int





  
  

  202
  
        Slice that seperates the anterior and weight bearing femoral cartilage.  





  
  

  203
  
    posteriorWBslice : int





  
  

  204
  
        Slice that seperates the weight bearing and posterior femoral cartilage. 





  
  

  205
  
    trochY : int





  
  

  206
  
        Slice that differentiates medial / lateral femur - trochlear notch Y component. 





  
  

  207
  
    femurLabel : int





  
  

  208
  
        Label that femur is in the segArray





  
  

  209
  
    medTibiaLabel : int





  
  

  210
  
        Label that medial tibia is in the segArray





  
  

  211
  
    latTibiaLabel : int





  
  

  212
  
        Label that lateral tibia is in the segArray





  
  

  213
  
    antFemurMask : int





  
  

  214
  
        Label anterior femur should be labeled in final segmentation. 





  
  

  215
  
    medWbFemurMask : int





  
  

  216
  
        Label medial weight bearing femur should be labeled in final segmentation.





  
  

  217
  
    latWbFemurMask : int





  
  

  218
  
        Label lateral weight bearing femur should be labeled in final segmentation. 





  
  

  219
  
    medPostFemurMask : int





  
  

  220
  
        Label medial posterior femur should be labeled in final segmentation. 





  
  

  221
  
    latPostFemurMask : int





  
  

  222
  
        Label lateral posterior femur should be labeled in final segmentation.





  
  

  223
  
    Returns





  
  

  224
  
    -------





  
  

  225
  
    final_segmentation : array





  
  

  226
  
        3D array with the updated segmentations - including weightbearing, medial/latera, anterior, and posterior. 





  
  

  227
  
    Notes





  
  

  228
  
    -----





  
  

  229
  
    





  
  

  230
  
    &#39;&#39;&#39;





  
  

  231
  
    





  
  

  232
  
    #array to store final segmentation





  
  

  233
  
    final_segmentation = np.zeros_like(segArray)





  
  

  234
  
    





  
  

  235
  
    #create masks for ant/wb/posterior femur





  
  

  236
  
    anterior_femur_mask = np.zeros_like(segArray)





  
  

  237
  
    anterior_femur_mask[:,:,:anteriorWBslice] = 1





  
  

  238
  





  
  

  239
  
    wb_femur_mask = np.zeros_like(segArray)





  
  

  240
  
    wb_femur_mask[:,:,anteriorWBslice:posteriorWBslice] = 1





  
  

  241
  





  
  

  242
  
    posterior_femur_mask = np.zeros_like(segArray)





  
  

  243
  
    posterior_femur_mask[:,:,posteriorWBslice:] = 1





  
  

  244
  
    





  
  

  245
  
    #create seg of just femur - and then break it into the sub-regions





  
  

  246
  
    femurSegArray = np.zeros_like(segArray)





  
  

  247
  
    femurSegArray[segArray==femurLabel] = 1





  
  

  248
  
    





  
  

  249
  
    #find the center of the medial/lateral tibia - use to distinguish M/L femur ROIs





  
  

  250
  
    locationMedialTibia = np.asarray(np.where(segArray==medTibiaLabel))





  
  

  251
  
    locationLateralTibia = np.asarray(np.where(segArray==latTibiaLabel))





  
  

  252
  
    





  
  

  253
  
    centerMedialTibia = locationMedialTibia.mean(axis=1)





  
  

  254
  
    centerLateralTibia = locationLateralTibia.mean(axis=1)





  
  

  255
  





  
  

  256
  
    med_femur_mask = np.zeros_like(segArray)





  
  

  257
  
    lat_femur_mask = np.zeros_like(segArray)





  
  

  258
  
    if centerMedialTibia[0] &gt; trochY:





  
  

  259
  
        med_femur_mask[trochY:,:,:] = 1





  
  

  260
  
        lat_femur_mask[:trochY,:,:] = 1





  
  

  261
  
    else:





  
  

  262
  
        med_femur_mask[:trochY,:,:] = 1





  
  

  263
  
        lat_femur_mask[trochY:,:,:] = 1





  
  

  264
  





  
  

  265
  
    final_segmentation[segArray!=femurLabel] = segArray[segArray!=femurLabel] 





  
  

  266
  
    final_segmentation += (femurSegArray * anterior_femur_mask) * antFemurMask





  
  

  267
  
    final_segmentation += (femurSegArray * wb_femur_mask * med_femur_mask) * medWbFemurMask





  
  

  268
  
    final_segmentation += (femurSegArray * wb_femur_mask * lat_femur_mask) * latWbFemurMask





  
  

  269
  
    final_segmentation += (femurSegArray * posterior_femur_mask * med_femur_mask) * medPostFemurMask





  
  

  270
  
    final_segmentation += (femurSegArray * posterior_femur_mask * lat_femur_mask) * latPostFemurMask





  
  

  271
  
    





  
  

  272
  
    return(final_segmentation)





  
  

  273
  





  
  

  274
  
def verify_and_correct_med_lat_tib_cart(





  
  

  275
  
    seg_array,  #sitk.GetArrayViewFromImage(seg)





  
  

  276
  
    tib_label=6,





  
  

  277
  
    med_tib_cart_label=2, 





  
  

  278
  
    lat_tib_cart_label=3,





  
  

  279
  
    ml_axis=0





  
  

  280
  
):





  
  

  281
  
    &#39;&#39;&#39;





  
  

  282
  
    Verify that the medial and lateral tibial cartilage are correctly labeled.





  
  

  283
  
    Parameters





  
  

  284
  
    ----------





  
  

  285
  
    seg_array : array





  
  

  286
  
        3D array with segmentation for the cartilage/bone regions.





  
  

  287
  
    tib_label : int





  
  

  288
  
        Label that tibial cartilage is in the seg_array





  
  

  289
  
    med_tib_cart_label : int





  
  

  290
  
        Label that medial tibial cartilage is in the seg_array





  
  

  291
  
    lat_tib_cart_label : int





  
  

  292
  
        Label that lateral tibial cartilage is in the seg_array





  
  

  293
  
    ml_axis : int





  
  

  294
  
        Medial/lateral axis of the acquired knee MRI.





  
  

  295
  
    





  
  

  296
  
    Returns





  
  

  297
  
    -------





  
  

  298
  
    seg_array : array





  
  

  299
  
        3D array with segmentation for the cartilage/bone regions.





  
  

  300
  
        The tibial cartilage regions will have been updated to ensure





  
  

  301
  
        all tib cart on med/lat sides are correctly classified.





  
  

  302
  
        





  
  

  303
  
    &#39;&#39;&#39;





  
  

  304
  
    #get binary array for tibia





  
  

  305
  
    array_tib = np.zeros_like(seg_array)





  
  

  306
  
    array_tib[seg_array == tib_label] = 1





  
  

  307
  
    #get binary array for tib cart





  
  

  308
  
    array_tib_cart = np.zeros_like(seg_array)





  
  

  309
  
    array_tib_cart[(seg_array == lat_tib_cart_label) + (seg_array == med_tib_cart_label)] = 1





  
  

  310
  





  
  

  311
  
    #get the locatons of med/lat cartilage &amp; get their centroids





  
  

  312
  
    med_cart_locs = np.asarray(np.where(seg_array == med_tib_cart_label))





  
  

  313
  
    lat_cart_locs = np.asarray(np.where(seg_array == lat_tib_cart_label))





  
  

  314
  
    middle_med_cart = med_cart_locs[ml_axis,:].mean()





  
  

  315
  
    middle_lat_cart = lat_cart_locs[ml_axis,:].mean()





  
  

  316
  





  
  

  317
  
    #get location of tibia to get centroid of tibial plateau





  
  

  318
  
    tib_locs = np.asarray(np.where(seg_array == tib_label))





  
  

  319
  
    middle_tib = tib_locs[ml_axis, :].mean()





  
  

  320
  
    center_tibia_slice = int(middle_tib)





  
  

  321
  





  
  

  322
  
    # infer the direction(s) for medial/lateral





  
  

  323
  
    med_direction = np.sign(middle_med_cart - middle_tib)





  
  

  324
  
    lat_direction = np.sign(middle_lat_cart - middle_tib)





  
  

  325
  
    if med_direction == lat_direction:





  
  

  326
  
        raise Exception(&#39;Middle of med and lat tibial cartilage on same side of centerline!&#39;)





  
  

  327
  





  
  

  328
  
    #create med/lat cartilage masks - binary for updating seg masks





  
  

  329
  
    med_tib_cart_mask = np.zeros_like(seg_array)





  
  

  330
  
    lat_tib_cart_mask = np.zeros_like(seg_array)





  
  

  331
  





  
  

  332
  
    if med_direction &gt; 0:





  
  

  333
  
        med_tib_cart_mask[center_tibia_slice:,...] = 1





  
  

  334
  
        lat_tib_cart_mask[:center_tibia_slice,...] = 1





  
  

  335
  
    elif med_direction &lt; 0:





  
  

  336
  
        med_tib_cart_mask[:center_tibia_slice,...] = 1





  
  

  337
  
        lat_tib_cart_mask[center_tibia_slice:,...] = 1





  
  

  338
  





  
  

  339
  
    # create new med/lat cartilage arrays 





  
  

  340
  
    new_med_cart_array = array_tib_cart * med_tib_cart_mask





  
  

  341
  
    new_lat_cart_array = array_tib_cart * lat_tib_cart_mask





  
  

  342
  





  
  

  343
  
    #make copy of original segmentation array &amp; update





  
  

  344
  
    # med/lat tibial cartilage labels





  
  

  345
  
    new_seg_array = seg_array.copy()





  
  

  346
  
    new_seg_array[new_med_cart_array == 1] = med_tib_cart_label





  
  

  347
  
    new_seg_array[new_lat_cart_array == 1] = lat_tib_cart_label





  
  

  348
  
    





  
  

  349
  
    return new_seg_array





  
  

  350
  





  
  

  351
  
def get_knee_segmentation_with_femur_subregions(seg_image,





  
  

  352
  
                                                fem_cart_label_idx=1,





  
  

  353
  
                                                wb_region_percent_dist=0.6,





  
  

  354
  
                                                # femur_label=1,





  
  

  355
  
                                                med_tibia_label=2,





  
  

  356
  
                                                lat_tibia_label=3,





  
  

  357
  
                                                ant_femur_mask=11,





  
  

  358
  
                                                med_wb_femur_mask=12,





  
  

  359
  
                                                lat_wb_femur_mask=13,





  
  

  360
  
                                                med_post_femur_mask=14,





  
  

  361
  
                                                lat_post_femur_mask=15,





  
  

  362
  
                                                verify_med_lat_tib_cart=True,





  
  

  363
  
                                                tibia_label=6,





  
  

  364
  
                                                ml_axis=0





  
  

  365
  
                                                ):





  
  

  366
  
    &#34;&#34;&#34;





  
  

  367
  
    Give seg image of knee. Return seg image with all sub-regions of femur included. 





  
  

  368
  





  
  

  369
  
    Parameters





  
  

  370
  
    ----------





  
  

  371
  
    seg_image : SimpleITK.Image





  
  

  372
  
        SimpleITK image of the segmentation to be processed. 





  
  

  373
  
    fem_cart_label_idx : int, optional





  
  

  374
  
        Label of femoral cartilage, by default 1





  
  

  375
  
    wb_region_percent_dist : float, optional





  
  

  376
  
        How large weightbearing region is (from not to posterior of condyles), by default 0.6





  
  

  377
  
    femur_label : int, optional





  
  

  378
  
        Seg label for the femur cartilage, by default 1





  
  

  379
  
    med_tibia_label : int, optional





  
  

  380
  
        Seg label for the medial tibia cartilage, by default 2





  
  

  381
  
    lat_tibia_label : int, optional





  
  

  382
  
        Seg label for the lateral tibia cartilage, by default 3





  
  

  383
  
    ant_femur_mask : int, optional





  
  

  384
  
        Seg label for the anterior femur region, by default 11





  
  

  385
  
    med_wb_femur_mask : int, optional





  
  

  386
  
        Seg label for medial weight-bearing femur, by default 12





  
  

  387
  
    lat_wb_femur_mask : int, optional





  
  

  388
  
        Seg label for lateral weight-bearing femur, by default 13





  
  

  389
  
    med_post_femur_mask : int, optional





  
  

  390
  
        Seg label for medial posterior femur, by default 14





  
  

  391
  
    lat_post_femur_mask : int, optional





  
  

  392
  
        Seg label for lateral posterior femur, by default 15





  
  

  393
  
    verify_med_lat_tib_cart : bool, optional





  
  

  394
  
        Whether to verify that medial and lateral tibial cartilage is on same side of centerline, by default True





  
  

  395
  
    tibia_label : int, optional





  
  

  396
  
        Seg label for the tibia, by default 6





  
  

  397
  
    ml_axis : int, optional





  
  

  398
  
        Medial/lateral axis of the acquired knee MRI, by default 0





  
  

  399
  





  
  

  400
  
    Returns





  
  

  401
  
    -------





  
  

  402
  
    SimpleITK.Image





  
  

  403
  
        Image of the new/updated segmentation





  
  

  404
  
    &#34;&#34;&#34;





  
  

  405
  
    troch_notch_y, troch_notch_x = getAnteriorOfWeightBearing(sitk.GetArrayViewFromImage(seg_image),





  
  

  406
  
                                                              femurIndex=fem_cart_label_idx)





  
  

  407
  
    loc_fem_z, loc_fem_y, loc_fem_x = np.where(sitk.GetArrayViewFromImage(seg_image) == fem_cart_label_idx)





  
  

  408
  
    post_femur_slice = np.max(loc_fem_x)





  
  

  409
  
    posterior_wb_slice = np.round((post_femur_slice - troch_notch_x) * wb_region_percent_dist + troch_notch_x).astype(int)





  
  

  410
  
    new_seg_array = getCartilageSubRegions(sitk.GetArrayViewFromImage(seg_image),





  
  

  411
  
                                           anteriorWBslice=troch_notch_x,





  
  

  412
  
                                           posteriorWBslice=posterior_wb_slice,





  
  

  413
  
                                           trochY=troch_notch_y,





  
  

  414
  
                                           femurLabel=fem_cart_label_idx,





  
  

  415
  
                                           medTibiaLabel=med_tibia_label,





  
  

  416
  
                                           latTibiaLabel=lat_tibia_label,





  
  

  417
  
                                           antFemurMask=ant_femur_mask,





  
  

  418
  
                                           medWbFemurMask=med_wb_femur_mask,





  
  

  419
  
                                           latWbFemurMask=lat_wb_femur_mask,





  
  

  420
  
                                           medPostFemurMask=med_post_femur_mask,





  
  

  421
  
                                           latPostFemurMask=lat_post_femur_mask





  
  

  422
  
                                           )





  
  

  423
  





  
  

  424
  
    if verify_med_lat_tib_cart:





  
  

  425
  
        new_seg_array = verify_and_correct_med_lat_tib_cart(new_seg_array,





  
  

  426
  
                                                            tib_label=tibia_label,





  
  

  427
  
                                                            med_tib_cart_label=med_tibia_label, 





  
  

  428
  
                                                            lat_tib_cart_label=lat_tibia_label,





  
  

  429
  
                                                            ml_axis=ml_axis) 





  
  

  430
  
    seg_label_image = sitk.GetImageFromArray(new_seg_array)





  
  

  431
  
    seg_label_image.CopyInformation(seg_image)





  
  

  432
  
    return seg_label_image</code></pre>





  
  

  433
  
</details>





  
  

  434
  
</section>





  
  

  435
  
<section>





  
  

  436
  
</section>





  
  

  437
  
<section>





  
  

  438
  
</section>





  
  

  439
  
<section>





  
  

  440
  
<h2 class="section-title" id="header-functions">Functions</h2>





  
  

  441
  
<dl>





  
  

  442
  
<dt id="pymskt.image.cartilage_processing.CofM"><code class="name flex">





  
  

  443
  
<span>def <span class="ident">CofM</span></span>(<span>array)</span>





  
  

  444
  
</code></dt>





  
  

  445
  
<dd>





  
  

  446
  
<div class="desc"><p>Get center of mass for a row of a binary 2D image.





  
  

  447
  
Parameters</p>





  
  

  448
  
<hr>





  
  

  449
  
<dl>





  
  

  450
  
<dt><strong><code>array</code></strong> :&ensp;<code>1D array</code></dt>





  
  

  451
  
<dd>Individual row of a 2D image.</dd>





  
  

  452
  
</dl>





  
  

  453
  
<h2 id="returns">Returns</h2>





  
  

  454
  
<dl>





  
  

  455
  
<dt><strong><code>centerPixels</code></strong></dt>





  
  

  456
  
<dd>Average location of 1s in the row</dd>





  
  

  457
  
</dl>





  
  

  458
  
<h2 id="notes">Notes</h2>





  
  

  459
  
<p>Calculates the average location of cartilage for the row of image being analyzed.





  
  

  460
  
Returns 0 if there are no pixels</p></div>





  
  

  461
  
<details class="source">





  
  

  462
  
<summary>





  
  

  463
  
<span>Expand source code</span>





  
  

  464
  
</summary>





  
  

  465
  
<pre><code class="python">def CofM(array):





  
  

  466
  
    &#39;&#39;&#39;





  
  

  467
  
    Get center of mass for a row of a binary 2D image. 





  
  

  468
  
    Parameters





  
  

  469
  
    ----------





  
  

  470
  
    array : 1D array





  
  

  471
  
        Individual row of a 2D image.  





  
  

  472
  
    Returns





  
  

  473
  
    -------





  
  

  474
  
    centerPixels : 





  
  

  475
  
        Average location of 1s in the row





  
  

  476
  
    Notes





  
  

  477
  
    -----





  
  

  478
  
    Calculates the average location of cartilage for the row of image being analyzed. 





  
  

  479
  
    Returns 0 if there are no pixels





  
  

  480
  
    





  
  

  481
  
    &#39;&#39;&#39;





  
  

  482
  
    pixels = np.where(array==1)





  
  

  483
  
    centerPixels = np.mean(pixels)





  
  

  484
  
    nans = np.isnan(centerPixels)





  
  

  485
  
    if nans == True:





  
  

  486
  
        centerPixels = 0





  
  

  487
  
    return(centerPixels)</code></pre>





  
  

  488
  
</details>





  
  

  489
  
</dd>





  
  

  490
  
<dt id="pymskt.image.cartilage_processing.absolute_CofM"><code class="name flex">





  
  

  491
  
<span>def <span class="ident">absolute_CofM</span></span>(<span>flattenedSeg)</span>





  
  

  492
  
</code></dt>





  
  

  493
  
<dd>





  
  

  494
  
<div class="desc"><p>Get absolute CofM of all the femoral cartilage pixels





  
  

  495
  
Parameters</p>





  
  

  496
  
<hr>





  
  

  497
  
<dl>





  
  

  498
  
<dt><strong><code>flattenedSeg</code></strong> :&ensp;<code>2D array</code></dt>





  
  

  499
  
<dd>Axial flattened, and filled in femoral cartilage segmentation.</dd>





  
  

  500
  
</dl>





  
  

  501
  
<h2 id="returns">Returns</h2>





  
  

  502
  
<dl>





  
  

  503
  
<dt><strong><code>centerX</code></strong></dt>





  
  

  504
  
<dd>The CofM in the X direction for the segmentation</dd>





  
  

  505
  
<dt><strong><code>centerY</code></strong></dt>





  
  

  506
  
<dd>The CofM in the Y direction for the segmentation</dd>





  
  

  507
  
</dl>





  
  

  508
  
<h2 id="notes">Notes</h2>





  
  

  509
  
<p>Get the x/y coordinates for the CofM for the whole flattened segmentation</p></div>





  
  

  510
  
<details class="source">





  
  

  511
  
<summary>





  
  

  512
  
<span>Expand source code</span>





  
  

  513
  
</summary>





  
  

  514
  
<pre><code class="python">def absolute_CofM(flattenedSeg):





  
  

  515
  
    &#39;&#39;&#39;





  
  

  516
  
    Get absolute CofM of all the femoral cartilage pixels





  
  

  517
  
    Parameters





  
  

  518
  
    ----------





  
  

  519
  
    flattenedSeg : 2D array





  
  

  520
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  521
  
    Returns





  
  

  522
  
    -------





  
  

  523
  
    centerX : 





  
  

  524
  
        The CofM in the X direction for the segmentation  





  
  

  525
  
    centerY : 





  
  

  526
  
        The CofM in the Y direction for the segmentation





  
  

  527
  
    Notes





  
  

  528
  
    -----





  
  

  529
  
    Get the x/y coordinates for the CofM for the whole flattened segmentation





  
  

  530
  
    





  
  

  531
  
    &#39;&#39;&#39;





  
  

  532
  
    femurPoints = np.where(flattenedSeg==1)





  
  

  533
  
    centerX = np.mean(femurPoints[0])





  
  

  534
  
    centerY = np.mean(femurPoints[1])





  
  

  535
  
    return(centerX, centerY)</code></pre>





  
  

  536
  
</details>





  
  

  537
  
</dd>





  
  

  538
  
<dt id="pymskt.image.cartilage_processing.findNotch"><code class="name flex">





  
  

  539
  
<span>def <span class="ident">findNotch</span></span>(<span>flattenedSeg, trochleaPositionX=1000)</span>





  
  

  540
  
</code></dt>





  
  

  541
  
<dd>





  
  

  542
  
<div class="desc"><p>Get the X Y position of the trochlear notch - where medial/lateral sides of the femur meet.





  
  

  543
  
Parameters</p>





  
  

  544
  
<hr>





  
  

  545
  
<dl>





  
  

  546
  
<dt><strong><code>flattenedSeg</code></strong> :&ensp;<code>2D array</code></dt>





  
  

  547
  
<dd>Axial flattened, and filled in femoral cartilage segmentation.</dd>





  
  

  548
  
</dl>





  
  

  549
  
<h2 id="returns">Returns</h2>





  
  

  550
  
<dl>





  
  

  551
  
<dt><strong><code>trochleaPositionY</code></strong></dt>





  
  

  552
  
<dd>Y position of trochlear notch</dd>





  
  

  553
  
<dt><strong><code>trochleaPositionX</code></strong></dt>





  
  

  554
  
<dd>X position of trochlear notch</dd>





  
  

  555
  
</dl>





  
  

  556
  
<h2 id="notes">Notes</h2>





  
  

  557
  
<p>Get the x/y coordinates for the trochlear notch. This is an iterative method that assumes things about the shape the





  
  

  558
  
femoral cartilage.</p></div>





  
  

  559
  
<details class="source">





  
  

  560
  
<summary>





  
  

  561
  
<span>Expand source code</span>





  
  

  562
  
</summary>





  
  

  563
  
<pre><code class="python">def findNotch(flattenedSeg, trochleaPositionX=1000):





  
  

  564
  
    &#39;&#39;&#39;





  
  

  565
  
    Get the X Y position of the trochlear notch - where medial/lateral sides of the femur meet. 





  
  

  566
  
    Parameters





  
  

  567
  
    ----------





  
  

  568
  
    flattenedSeg : 2D array





  
  

  569
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  570
  
    Returns





  
  

  571
  
    -------





  
  

  572
  
    trochleaPositionY : 





  
  

  573
  
        Y position of trochlear notch  





  
  

  574
  
    trochleaPositionX : 





  
  

  575
  
        X position of trochlear notch





  
  

  576
  
    Notes





  
  

  577
  
    -----





  
  

  578
  
    Get the x/y coordinates for the trochlear notch. This is an iterative method that assumes things about the shape the





  
  

  579
  
    femoral cartilage.





  
  

  580
  
    





  
  

  581
  
    &#39;&#39;&#39;





  
  

  582
  
    # Goal is to find the most anterior point that is between the medial/lateral condyles





  
  

  583
  





  
  

  584
  
    # First guess at the troch notch in the 1st axis (med/lat axis) is the location with the smallest value for





  
  

  585
  
    # the 2nd axis CofM. This is because in axis 1, negative is anterior and we expect the most anterior CofM should 





  
  

  586
  
    # roughly align with the trochlear notch.





  
  

  587
  
    y_CofM = get_y_CofM(flattenedSeg)





  
  

  588
  
    first_guess = y_CofM[np.argmin(y_CofM[:,1]), 0]





  
  

  589
  
    # the second guess is just the CofM of the whole cartilage. 





  
  

  590
  
    centerX, centerY = absolute_CofM(flattenedSeg)





  
  

  591
  
    second_guess = centerX





  
  

  592
  





  
  

  593
  
    # We use the 2 guesses to help define a search space for the trochlear notch.





  
  

  594
  
    min_search = int(np.min((first_guess,second_guess))-20)





  
  

  595
  
    max_search = int(np.max((first_guess,second_guess))+20)





  
  

  596
  





  
  

  597
  
    # now, we iterate over all of the rows (axis 1) of the search space (moving in the medial/lateral direction)





  
  

  598
  
    # we are looking for the row where the most posterior point (back of femur) is furthest anterior (notch). 





  
  

  599
  
    for y in range(min_search, max_search):





  
  

  600
  
        # At each row, we find most posterior pixel labeled as cartilage. 





  
  

  601
  
        try:





  
  

  602
  
            trochleaPosition_test = np.max(np.where(flattenedSeg[y,:]==1))





  
  

  603
  
        except ValueError:





  
  

  604
  
            # if there is no cartilage we&#39;ll get a ValueError exception. 





  
  

  605
  
            # in that case, set this value to be the max it can be (the size of the first axis)





  
  

  606
  
            trochleaPosition_test = flattenedSeg.shape[1]





  
  

  607
  
        # if the most posterior point for this row is more anterior than the current trochleaPositionX,





  
  

  608
  
        # then update this to be the new trochlear notch.





  
  

  609
  
        if trochleaPosition_test &lt; trochleaPositionX:





  
  

  610
  
            trochleaPositionX = trochleaPosition_test





  
  

  611
  
            trochleaPositionY = y





  
  

  612
  





  
  

  613
  
    return(trochleaPositionY, trochleaPositionX+1)</code></pre>





  
  

  614
  
</details>





  
  

  615
  
</dd>





  
  

  616
  
<dt id="pymskt.image.cartilage_processing.getAnteriorOfWeightBearing"><code class="name flex">





  
  

  617
  
<span>def <span class="ident">getAnteriorOfWeightBearing</span></span>(<span>segArray, femurIndex=1)</span>





  
  

  618
  
</code></dt>





  
  

  619
  
<dd>





  
  

  620
  
<div class="desc"><p>Prepare full segmentation and extract the trochlear notch location.





  
  

  621
  
Parameters</p>





  
  

  622
  
<hr>





  
  

  623
  
<dl>





  
  

  624
  
<dt><strong><code>flattenedSeg</code></strong> :&ensp;<code>2D array</code></dt>





  
  

  625
  
<dd>Axial flattened, and filled in femoral cartilage segmentation.</dd>





  
  

  626
  
<dt><strong><code>femurIndex</code></strong> :&ensp;<code>int</code></dt>





  
  

  627
  
<dd>Index of the label used to localize the femur in the array.</dd>





  
  

  628
  
</dl>





  
  

  629
  
<h2 id="returns">Returns</h2>





  
  

  630
  
<dl>





  
  

  631
  
<dt><strong><code>trochleaPositionY</code></strong></dt>





  
  

  632
  
<dd>Y position of trochlear notch</dd>





  
  

  633
  
<dt><strong><code>trochleaPositionX</code></strong></dt>





  
  

  634
  
<dd>X position of trochlear notch</dd>





  
  

  635
  
</dl>





  
  

  636
  
<h2 id="notes">Notes</h2>





  
  

  637
  
<p>Get the x/y coordinates for the trochlear notch. This is an iterative method that assumes things about the shape the femoral cartilage.





  
  

  638
  
First flatten and fill any holes in the segmentation.</p></div>





  
  

  639
  
<details class="source">





  
  

  640
  
<summary>





  
  

  641
  
<span>Expand source code</span>





  
  

  642
  
</summary>





  
  

  643
  
<pre><code class="python">def getAnteriorOfWeightBearing(segArray, femurIndex=1):





  
  

  644
  
    &#39;&#39;&#39;





  
  

  645
  
    Prepare full segmentation and extract the trochlear notch location. 





  
  

  646
  
    Parameters





  
  

  647
  
    ----------





  
  

  648
  
    flattenedSeg : 2D array





  
  

  649
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  650
  
    femurIndex : int





  
  

  651
  
        Index of the label used to localize the femur in the array. 





  
  

  652
  
    Returns





  
  

  653
  
    -------





  
  

  654
  
    trochleaPositionY : 





  
  

  655
  
        Y position of trochlear notch  





  
  

  656
  
    trochleaPositionX : 





  
  

  657
  
        X position of trochlear notch





  
  

  658
  
    Notes





  
  

  659
  
    -----





  
  

  660
  
    Get the x/y coordinates for the trochlear notch. This is an iterative method that assumes things about the shape the femoral cartilage. 





  
  

  661
  
    First flatten and fill any holes in the segmentation. 





  
  

  662
  
    





  
  

  663
  
    &#39;&#39;&#39;





  
  

  664
  





  
  

  665
  
    femurSegmentation = np.zeros_like(segArray)





  
  

  666
  
    femurSegmentation[segArray == femurIndex] = 1





  
  

  667
  
    flattenedSegmentation = np.amax(femurSegmentation, axis=1)





  
  

  668
  
    flattened_seg_filled = ndi.binary_fill_holes(flattenedSegmentation)





  
  

  669
  
    trochY, trochX = findNotch(flattened_seg_filled)





  
  

  670
  
    return(trochY, trochX)</code></pre>





  
  

  671
  
</details>





  
  

  672
  
</dd>





  
  

  673
  
<dt id="pymskt.image.cartilage_processing.getCartilageSubRegions"><code class="name flex">





  
  

  674
  
<span>def <span class="ident">getCartilageSubRegions</span></span>(<span>segArray, anteriorWBslice, posteriorWBslice, trochY, femurLabel=1, medTibiaLabel=2, latTibiaLabel=3, antFemurMask=5, medWbFemurMask=6, latWbFemurMask=7, medPostFemurMask=8, latPostFemurMask=9)</span>





  
  

  675
  
</code></dt>





  
  

  676
  
<dd>





  
  

  677
  
<div class="desc"><p>Take cartilage segmentation, and decompose femoral cartilage into subregions of interest.<br>





  
  

  678
  
Parameters</p>





  
  

  679
  
<hr>





  
  

  680
  
<dl>





  
  

  681
  
<dt><strong><code>segArray</code></strong> :&ensp;<code>array</code></dt>





  
  

  682
  
<dd>3D array with segmentation for the cartialge regions.</dd>





  
  

  683
  
<dt><strong><code>anteriorWBslice</code></strong> :&ensp;<code>int</code></dt>





  
  

  684
  
<dd>Slice that seperates the anterior and weight bearing femoral cartilage.</dd>





  
  

  685
  
<dt><strong><code>posteriorWBslice</code></strong> :&ensp;<code>int</code></dt>





  
  

  686
  
<dd>Slice that seperates the weight bearing and posterior femoral cartilage.</dd>





  
  

  687
  
<dt><strong><code>trochY</code></strong> :&ensp;<code>int</code></dt>





  
  

  688
  
<dd>Slice that differentiates medial / lateral femur - trochlear notch Y component.</dd>





  
  

  689
  
<dt><strong><code>femurLabel</code></strong> :&ensp;<code>int</code></dt>





  
  

  690
  
<dd>Label that femur is in the segArray</dd>





  
  

  691
  
<dt><strong><code>medTibiaLabel</code></strong> :&ensp;<code>int</code></dt>





  
  

  692
  
<dd>Label that medial tibia is in the segArray</dd>





  
  

  693
  
<dt><strong><code>latTibiaLabel</code></strong> :&ensp;<code>int</code></dt>





  
  

  694
  
<dd>Label that lateral tibia is in the segArray</dd>





  
  

  695
  
<dt><strong><code>antFemurMask</code></strong> :&ensp;<code>int</code></dt>





  
  

  696
  
<dd>Label anterior femur should be labeled in final segmentation.</dd>





  
  

  697
  
<dt><strong><code>medWbFemurMask</code></strong> :&ensp;<code>int</code></dt>





  
  

  698
  
<dd>Label medial weight bearing femur should be labeled in final segmentation.</dd>





  
  

  699
  
<dt><strong><code>latWbFemurMask</code></strong> :&ensp;<code>int</code></dt>





  
  

  700
  
<dd>Label lateral weight bearing femur should be labeled in final segmentation.</dd>





  
  

  701
  
<dt><strong><code>medPostFemurMask</code></strong> :&ensp;<code>int</code></dt>





  
  

  702
  
<dd>Label medial posterior femur should be labeled in final segmentation.</dd>





  
  

  703
  
<dt><strong><code>latPostFemurMask</code></strong> :&ensp;<code>int</code></dt>





  
  

  704
  
<dd>Label lateral posterior femur should be labeled in final segmentation.</dd>





  
  

  705
  
</dl>





  
  

  706
  
<h2 id="returns">Returns</h2>





  
  

  707
  
<dl>





  
  

  708
  
<dt><strong><code>final_segmentation</code></strong> :&ensp;<code>array</code></dt>





  
  

  709
  
<dd>3D array with the updated segmentations - including weightbearing, medial/latera, anterior, and posterior.</dd>





  
  

  710
  
</dl>





  
  

  711
  
<h2 id="notes">Notes</h2></div>





  
  

  712
  
<details class="source">





  
  

  713
  
<summary>





  
  

  714
  
<span>Expand source code</span>





  
  

  715
  
</summary>





  
  

  716
  
<pre><code class="python">def getCartilageSubRegions(segArray, anteriorWBslice, posteriorWBslice, trochY,





  
  

  717
  
                           femurLabel=1, medTibiaLabel=2, latTibiaLabel=3, antFemurMask=5, 





  
  

  718
  
                           medWbFemurMask=6, latWbFemurMask=7, medPostFemurMask=8, latPostFemurMask=9):





  
  

  719
  
    &#39;&#39;&#39;





  
  

  720
  
    Take cartilage segmentation, and decompose femoral cartilage into subregions of interest.  





  
  

  721
  
    Parameters





  
  

  722
  
    ----------





  
  

  723
  
    segArray : array





  
  

  724
  
        3D array with segmentation for the cartialge regions. 





  
  

  725
  
    anteriorWBslice : int





  
  

  726
  
        Slice that seperates the anterior and weight bearing femoral cartilage.  





  
  

  727
  
    posteriorWBslice : int





  
  

  728
  
        Slice that seperates the weight bearing and posterior femoral cartilage. 





  
  

  729
  
    trochY : int





  
  

  730
  
        Slice that differentiates medial / lateral femur - trochlear notch Y component. 





  
  

  731
  
    femurLabel : int





  
  

  732
  
        Label that femur is in the segArray





  
  

  733
  
    medTibiaLabel : int





  
  

  734
  
        Label that medial tibia is in the segArray





  
  

  735
  
    latTibiaLabel : int





  
  

  736
  
        Label that lateral tibia is in the segArray





  
  

  737
  
    antFemurMask : int





  
  

  738
  
        Label anterior femur should be labeled in final segmentation. 





  
  

  739
  
    medWbFemurMask : int





  
  

  740
  
        Label medial weight bearing femur should be labeled in final segmentation.





  
  

  741
  
    latWbFemurMask : int





  
  

  742
  
        Label lateral weight bearing femur should be labeled in final segmentation. 





  
  

  743
  
    medPostFemurMask : int





  
  

  744
  
        Label medial posterior femur should be labeled in final segmentation. 





  
  

  745
  
    latPostFemurMask : int





  
  

  746
  
        Label lateral posterior femur should be labeled in final segmentation.





  
  

  747
  
    Returns





  
  

  748
  
    -------





  
  

  749
  
    final_segmentation : array





  
  

  750
  
        3D array with the updated segmentations - including weightbearing, medial/latera, anterior, and posterior. 





  
  

  751
  
    Notes





  
  

  752
  
    -----





  
  

  753
  
    





  
  

  754
  
    &#39;&#39;&#39;





  
  

  755
  
    





  
  

  756
  
    #array to store final segmentation





  
  

  757
  
    final_segmentation = np.zeros_like(segArray)





  
  

  758
  
    





  
  

  759
  
    #create masks for ant/wb/posterior femur





  
  

  760
  
    anterior_femur_mask = np.zeros_like(segArray)





  
  

  761
  
    anterior_femur_mask[:,:,:anteriorWBslice] = 1





  
  

  762
  





  
  

  763
  
    wb_femur_mask = np.zeros_like(segArray)





  
  

  764
  
    wb_femur_mask[:,:,anteriorWBslice:posteriorWBslice] = 1





  
  

  765
  





  
  

  766
  
    posterior_femur_mask = np.zeros_like(segArray)





  
  

  767
  
    posterior_femur_mask[:,:,posteriorWBslice:] = 1





  
  

  768
  
    





  
  

  769
  
    #create seg of just femur - and then break it into the sub-regions





  
  

  770
  
    femurSegArray = np.zeros_like(segArray)





  
  

  771
  
    femurSegArray[segArray==femurLabel] = 1





  
  

  772
  
    





  
  

  773
  
    #find the center of the medial/lateral tibia - use to distinguish M/L femur ROIs





  
  

  774
  
    locationMedialTibia = np.asarray(np.where(segArray==medTibiaLabel))





  
  

  775
  
    locationLateralTibia = np.asarray(np.where(segArray==latTibiaLabel))





  
  

  776
  
    





  
  

  777
  
    centerMedialTibia = locationMedialTibia.mean(axis=1)





  
  

  778
  
    centerLateralTibia = locationLateralTibia.mean(axis=1)





  
  

  779
  





  
  

  780
  
    med_femur_mask = np.zeros_like(segArray)





  
  

  781
  
    lat_femur_mask = np.zeros_like(segArray)





  
  

  782
  
    if centerMedialTibia[0] &gt; trochY:





  
  

  783
  
        med_femur_mask[trochY:,:,:] = 1





  
  

  784
  
        lat_femur_mask[:trochY,:,:] = 1





  
  

  785
  
    else:





  
  

  786
  
        med_femur_mask[:trochY,:,:] = 1





  
  

  787
  
        lat_femur_mask[trochY:,:,:] = 1





  
  

  788
  





  
  

  789
  
    final_segmentation[segArray!=femurLabel] = segArray[segArray!=femurLabel] 





  
  

  790
  
    final_segmentation += (femurSegArray * anterior_femur_mask) * antFemurMask





  
  

  791
  
    final_segmentation += (femurSegArray * wb_femur_mask * med_femur_mask) * medWbFemurMask





  
  

  792
  
    final_segmentation += (femurSegArray * wb_femur_mask * lat_femur_mask) * latWbFemurMask





  
  

  793
  
    final_segmentation += (femurSegArray * posterior_femur_mask * med_femur_mask) * medPostFemurMask





  
  

  794
  
    final_segmentation += (femurSegArray * posterior_femur_mask * lat_femur_mask) * latPostFemurMask





  
  

  795
  
    





  
  

  796
  
    return(final_segmentation)</code></pre>





  
  

  797
  
</details>





  
  

  798
  
</dd>





  
  

  799
  
<dt id="pymskt.image.cartilage_processing.get_knee_segmentation_with_femur_subregions"><code class="name flex">





  
  

  800
  
<span>def <span class="ident">get_knee_segmentation_with_femur_subregions</span></span>(<span>seg_image, fem_cart_label_idx=1, wb_region_percent_dist=0.6, med_tibia_label=2, lat_tibia_label=3, ant_femur_mask=11, med_wb_femur_mask=12, lat_wb_femur_mask=13, med_post_femur_mask=14, lat_post_femur_mask=15, verify_med_lat_tib_cart=True, tibia_label=6, ml_axis=0)</span>





  
  

  801
  
</code></dt>





  
  

  802
  
<dd>





  
  

  803
  
<div class="desc"><p>Give seg image of knee. Return seg image with all sub-regions of femur included. </p>





  
  

  804
  
<h2 id="parameters">Parameters</h2>





  
  

  805
  
<dl>





  
  

  806
  
<dt><strong><code>seg_image</code></strong> :&ensp;<code>SimpleITK.Image</code></dt>





  
  

  807
  
<dd>SimpleITK image of the segmentation to be processed.</dd>





  
  

  808
  
<dt><strong><code>fem_cart_label_idx</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  809
  
<dd>Label of femoral cartilage, by default 1</dd>





  
  

  810
  
<dt><strong><code>wb_region_percent_dist</code></strong> :&ensp;<code>float</code>, optional</dt>





  
  

  811
  
<dd>How large weightbearing region is (from not to posterior of condyles), by default 0.6</dd>





  
  

  812
  
<dt><strong><code>femur_label</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  813
  
<dd>Seg label for the femur cartilage, by default 1</dd>





  
  

  814
  
<dt><strong><code>med_tibia_label</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  815
  
<dd>Seg label for the medial tibia cartilage, by default 2</dd>





  
  

  816
  
<dt><strong><code>lat_tibia_label</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  817
  
<dd>Seg label for the lateral tibia cartilage, by default 3</dd>





  
  

  818
  
<dt><strong><code>ant_femur_mask</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  819
  
<dd>Seg label for the anterior femur region, by default 11</dd>





  
  

  820
  
<dt><strong><code>med_wb_femur_mask</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  821
  
<dd>Seg label for medial weight-bearing femur, by default 12</dd>





  
  

  822
  
<dt><strong><code>lat_wb_femur_mask</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  823
  
<dd>Seg label for lateral weight-bearing femur, by default 13</dd>





  
  

  824
  
<dt><strong><code>med_post_femur_mask</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  825
  
<dd>Seg label for medial posterior femur, by default 14</dd>





  
  

  826
  
<dt><strong><code>lat_post_femur_mask</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  827
  
<dd>Seg label for lateral posterior femur, by default 15</dd>





  
  

  828
  
<dt><strong><code>verify_med_lat_tib_cart</code></strong> :&ensp;<code>bool</code>, optional</dt>





  
  

  829
  
<dd>Whether to verify that medial and lateral tibial cartilage is on same side of centerline, by default True</dd>





  
  

  830
  
<dt><strong><code>tibia_label</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  831
  
<dd>Seg label for the tibia, by default 6</dd>





  
  

  832
  
<dt><strong><code>ml_axis</code></strong> :&ensp;<code>int</code>, optional</dt>





  
  

  833
  
<dd>Medial/lateral axis of the acquired knee MRI, by default 0</dd>





  
  

  834
  
</dl>





  
  

  835
  
<h2 id="returns">Returns</h2>





  
  

  836
  
<dl>





  
  

  837
  
<dt><code>SimpleITK.Image</code></dt>





  
  

  838
  
<dd>Image of the new/updated segmentation</dd>





  
  

  839
  
</dl></div>





  
  

  840
  
<details class="source">





  
  

  841
  
<summary>





  
  

  842
  
<span>Expand source code</span>





  
  

  843
  
</summary>





  
  

  844
  
<pre><code class="python">def get_knee_segmentation_with_femur_subregions(seg_image,





  
  

  845
  
                                                fem_cart_label_idx=1,





  
  

  846
  
                                                wb_region_percent_dist=0.6,





  
  

  847
  
                                                # femur_label=1,





  
  

  848
  
                                                med_tibia_label=2,





  
  

  849
  
                                                lat_tibia_label=3,





  
  

  850
  
                                                ant_femur_mask=11,





  
  

  851
  
                                                med_wb_femur_mask=12,





  
  

  852
  
                                                lat_wb_femur_mask=13,





  
  

  853
  
                                                med_post_femur_mask=14,





  
  

  854
  
                                                lat_post_femur_mask=15,





  
  

  855
  
                                                verify_med_lat_tib_cart=True,





  
  

  856
  
                                                tibia_label=6,





  
  

  857
  
                                                ml_axis=0





  
  

  858
  
                                                ):





  
  

  859
  
    &#34;&#34;&#34;





  
  

  860
  
    Give seg image of knee. Return seg image with all sub-regions of femur included. 





  
  

  861
  





  
  

  862
  
    Parameters





  
  

  863
  
    ----------





  
  

  864
  
    seg_image : SimpleITK.Image





  
  

  865
  
        SimpleITK image of the segmentation to be processed. 





  
  

  866
  
    fem_cart_label_idx : int, optional





  
  

  867
  
        Label of femoral cartilage, by default 1





  
  

  868
  
    wb_region_percent_dist : float, optional





  
  

  869
  
        How large weightbearing region is (from not to posterior of condyles), by default 0.6





  
  

  870
  
    femur_label : int, optional





  
  

  871
  
        Seg label for the femur cartilage, by default 1





  
  

  872
  
    med_tibia_label : int, optional





  
  

  873
  
        Seg label for the medial tibia cartilage, by default 2





  
  

  874
  
    lat_tibia_label : int, optional





  
  

  875
  
        Seg label for the lateral tibia cartilage, by default 3





  
  

  876
  
    ant_femur_mask : int, optional





  
  

  877
  
        Seg label for the anterior femur region, by default 11





  
  

  878
  
    med_wb_femur_mask : int, optional





  
  

  879
  
        Seg label for medial weight-bearing femur, by default 12





  
  

  880
  
    lat_wb_femur_mask : int, optional





  
  

  881
  
        Seg label for lateral weight-bearing femur, by default 13





  
  

  882
  
    med_post_femur_mask : int, optional





  
  

  883
  
        Seg label for medial posterior femur, by default 14





  
  

  884
  
    lat_post_femur_mask : int, optional





  
  

  885
  
        Seg label for lateral posterior femur, by default 15





  
  

  886
  
    verify_med_lat_tib_cart : bool, optional





  
  

  887
  
        Whether to verify that medial and lateral tibial cartilage is on same side of centerline, by default True





  
  

  888
  
    tibia_label : int, optional





  
  

  889
  
        Seg label for the tibia, by default 6





  
  

  890
  
    ml_axis : int, optional





  
  

  891
  
        Medial/lateral axis of the acquired knee MRI, by default 0





  
  

  892
  





  
  

  893
  
    Returns





  
  

  894
  
    -------





  
  

  895
  
    SimpleITK.Image





  
  

  896
  
        Image of the new/updated segmentation





  
  

  897
  
    &#34;&#34;&#34;





  
  

  898
  
    troch_notch_y, troch_notch_x = getAnteriorOfWeightBearing(sitk.GetArrayViewFromImage(seg_image),





  
  

  899
  
                                                              femurIndex=fem_cart_label_idx)





  
  

  900
  
    loc_fem_z, loc_fem_y, loc_fem_x = np.where(sitk.GetArrayViewFromImage(seg_image) == fem_cart_label_idx)





  
  

  901
  
    post_femur_slice = np.max(loc_fem_x)





  
  

  902
  
    posterior_wb_slice = np.round((post_femur_slice - troch_notch_x) * wb_region_percent_dist + troch_notch_x).astype(int)





  
  

  903
  
    new_seg_array = getCartilageSubRegions(sitk.GetArrayViewFromImage(seg_image),





  
  

  904
  
                                           anteriorWBslice=troch_notch_x,





  
  

  905
  
                                           posteriorWBslice=posterior_wb_slice,





  
  

  906
  
                                           trochY=troch_notch_y,





  
  

  907
  
                                           femurLabel=fem_cart_label_idx,





  
  

  908
  
                                           medTibiaLabel=med_tibia_label,





  
  

  909
  
                                           latTibiaLabel=lat_tibia_label,





  
  

  910
  
                                           antFemurMask=ant_femur_mask,





  
  

  911
  
                                           medWbFemurMask=med_wb_femur_mask,





  
  

  912
  
                                           latWbFemurMask=lat_wb_femur_mask,





  
  

  913
  
                                           medPostFemurMask=med_post_femur_mask,





  
  

  914
  
                                           latPostFemurMask=lat_post_femur_mask





  
  

  915
  
                                           )





  
  

  916
  





  
  

  917
  
    if verify_med_lat_tib_cart:





  
  

  918
  
        new_seg_array = verify_and_correct_med_lat_tib_cart(new_seg_array,





  
  

  919
  
                                                            tib_label=tibia_label,





  
  

  920
  
                                                            med_tib_cart_label=med_tibia_label, 





  
  

  921
  
                                                            lat_tib_cart_label=lat_tibia_label,





  
  

  922
  
                                                            ml_axis=ml_axis) 





  
  

  923
  
    seg_label_image = sitk.GetImageFromArray(new_seg_array)





  
  

  924
  
    seg_label_image.CopyInformation(seg_image)





  
  

  925
  
    return seg_label_image</code></pre>





  
  

  926
  
</details>





  
  

  927
  
</dd>





  
  

  928
  
<dt id="pymskt.image.cartilage_processing.get_y_CofM"><code class="name flex">





  
  

  929
  
<span>def <span class="ident">get_y_CofM</span></span>(<span>flattenedSeg)</span>





  
  

  930
  
</code></dt>





  
  

  931
  
<dd>





  
  

  932
  
<div class="desc"><p>Get CofM of femoral cartilage for each row of the flattened segmentation.





  
  

  933
  
Parameters</p>





  
  

  934
  
<hr>





  
  

  935
  
<dl>





  
  

  936
  
<dt><strong><code>flattenedSeg</code></strong> :&ensp;<code>2D array</code></dt>





  
  

  937
  
<dd>Axial flattened, and filled in femoral cartilage segmentation.</dd>





  
  

  938
  
</dl>





  
  

  939
  
<h2 id="returns">Returns</h2>





  
  

  940
  
<dl>





  
  

  941
  
<dt><strong><code>yCofM</code></strong></dt>





  
  

  942
  
<dd>Find the CofM for each row of the image.</dd>





  
  

  943
  
</dl>





  
  

  944
  
<h2 id="notes">Notes</h2>





  
  

  945
  
<p>Get the x/y coordinates for the CofM for each row of the flattened segmentation.</p></div>





  
  

  946
  
<details class="source">





  
  

  947
  
<summary>





  
  

  948
  
<span>Expand source code</span>





  
  

  949
  
</summary>





  
  

  950
  
<pre><code class="python">def get_y_CofM(flattenedSeg):





  
  

  951
  
    &#39;&#39;&#39;





  
  

  952
  
    Get CofM of femoral cartilage for each row of the flattened segmentation. 





  
  

  953
  
    Parameters





  
  

  954
  
    ----------





  
  

  955
  
    flattenedSeg : 2D array





  
  

  956
  
        Axial flattened, and filled in femoral cartilage segmentation.  





  
  

  957
  
    Returns





  
  

  958
  
    -------





  
  

  959
  
    yCofM : 





  
  

  960
  
        Find the CofM for each row of the image. 





  
  

  961
  
    Notes





  
  

  962
  
    -----





  
  

  963
  
    Get the x/y coordinates for the CofM for each row of the flattened segmentation. 





  
  

  964
  
    





  
  

  965
  
    &#39;&#39;&#39;





  
  

  966
  
    locationFemur = np.where(flattenedSeg==1)





  
  

  967
  
    yCofM = np.zeros((flattenedSeg.shape[0], 2), dtype=int)





  
  

  968
  
    





  
  

  969
  
    # only calculate for rows with cartilage. 





  
  

  970
  
    minRow = np.min(locationFemur[0])





  
  

  971
  
    maxRow = np.max(locationFemur[0])





  
  

  972
  
    





  
  

  973
  
    # iterate over rows of image, get CofM, store CofM for row. 





  
  

  974
  
    for x in range(minRow, maxRow):





  
  

  975
  
        yCofM[x, 0] = x #store the x-coordinate (row) we calcualted CofM for. 





  
  

  976
  
        yCofM[x, 1] = int(CofM(flattenedSeg[x, :])) # store the CofM value (make it an integer for indexing)





  
  

  977
  
    yCofM = yCofM[minRow+10:maxRow-10,:] # remove 10 most medial and most lateral pixels of femoral cartilage. 





  
  

  978
  
    return(yCofM) </code></pre>





  
  

  979
  
</details>





  
  

  980
  
</dd>





  
  

  981
  
<dt id="pymskt.image.cartilage_processing.verify_and_correct_med_lat_tib_cart"><code class="name flex">





  
  

  982
  
<span>def <span class="ident">verify_and_correct_med_lat_tib_cart</span></span>(<span>seg_array, tib_label=6, med_tib_cart_label=2, lat_tib_cart_label=3, ml_axis=0)</span>





  
  

  983
  
</code></dt>





  
  

  984
  
<dd>





  
  

  985
  
<div class="desc"><p>Verify that the medial and lateral tibial cartilage are correctly labeled.





  
  

  986
  
Parameters</p>





  
  

  987
  
<hr>





  
  

  988
  
<dl>





  
  

  989
  
<dt><strong><code>seg_array</code></strong> :&ensp;<code>array</code></dt>





  
  

  990
  
<dd>3D array with segmentation for the cartilage/bone regions.</dd>





  
  

  991
  
<dt><strong><code>tib_label</code></strong> :&ensp;<code>int</code></dt>





  
  

  992
  
<dd>Label that tibial cartilage is in the seg_array</dd>





  
  

  993
  
<dt><strong><code>med_tib_cart_label</code></strong> :&ensp;<code>int</code></dt>





  
  

  994
  
<dd>Label that medial tibial cartilage is in the seg_array</dd>





  
  

  995
  
<dt><strong><code>lat_tib_cart_label</code></strong> :&ensp;<code>int</code></dt>





  
  

  996
  
<dd>Label that lateral tibial cartilage is in the seg_array</dd>





  
  

  997
  
<dt><strong><code>ml_axis</code></strong> :&ensp;<code>int</code></dt>





  
  

  998
  
<dd>Medial/lateral axis of the acquired knee MRI.</dd>





  
  

  999
  
</dl>





  
  

  1000
  
<h2 id="returns">Returns</h2>





  
  

  1001
  
<dl>





  
  

  1002
  
<dt><strong><code>seg_array</code></strong> :&ensp;<code>array</code></dt>





  
  

  1003
  
<dd>3D array with segmentation for the cartilage/bone regions.





  
  

  1004
  
The tibial cartilage regions will have been updated to ensure





  
  

  1005
  
all tib cart on med/lat sides are correctly classified.</dd>





  
  

  1006
  
</dl></div>





  
  

  1007
  
<details class="source">





  
  

  1008
  
<summary>





  
  

  1009
  
<span>Expand source code</span>





  
  

  1010
  
</summary>





  
  

  1011
  
<pre><code class="python">def verify_and_correct_med_lat_tib_cart(





  
  

  1012
  
    seg_array,  #sitk.GetArrayViewFromImage(seg)





  
  

  1013
  
    tib_label=6,





  
  

  1014
  
    med_tib_cart_label=2, 





  
  

  1015
  
    lat_tib_cart_label=3,





  
  

  1016
  
    ml_axis=0





  
  

  1017
  
):





  
  

  1018
  
    &#39;&#39;&#39;





  
  

  1019
  
    Verify that the medial and lateral tibial cartilage are correctly labeled.





  
  

  1020
  
    Parameters





  
  

  1021
  
    ----------





  
  

  1022
  
    seg_array : array





  
  

  1023
  
        3D array with segmentation for the cartilage/bone regions.





  
  

  1024
  
    tib_label : int





  
  

  1025
  
        Label that tibial cartilage is in the seg_array





  
  

  1026
  
    med_tib_cart_label : int





  
  

  1027
  
        Label that medial tibial cartilage is in the seg_array





  
  

  1028
  
    lat_tib_cart_label : int





  
  

  1029
  
        Label that lateral tibial cartilage is in the seg_array





  
  

  1030
  
    ml_axis : int





  
  

  1031
  
        Medial/lateral axis of the acquired knee MRI.





  
  

  1032
  
    





  
  

  1033
  
    Returns





  
  

  1034
  
    -------





  
  

  1035
  
    seg_array : array





  
  

  1036
  
        3D array with segmentation for the cartilage/bone regions.





  
  

  1037
  
        The tibial cartilage regions will have been updated to ensure





  
  

  1038
  
        all tib cart on med/lat sides are correctly classified.





  
  

  1039
  
        





  
  

  1040
  
    &#39;&#39;&#39;





  
  

  1041
  
    #get binary array for tibia





  
  

  1042
  
    array_tib = np.zeros_like(seg_array)





  
  

  1043
  
    array_tib[seg_array == tib_label] = 1





  
  

  1044
  
    #get binary array for tib cart





  
  

  1045
  
    array_tib_cart = np.zeros_like(seg_array)





  
  

  1046
  
    array_tib_cart[(seg_array == lat_tib_cart_label) + (seg_array == med_tib_cart_label)] = 1





  
  

  1047
  





  
  

  1048
  
    #get the locatons of med/lat cartilage &amp; get their centroids





  
  

  1049
  
    med_cart_locs = np.asarray(np.where(seg_array == med_tib_cart_label))





  
  

  1050
  
    lat_cart_locs = np.asarray(np.where(seg_array == lat_tib_cart_label))





  
  

  1051
  
    middle_med_cart = med_cart_locs[ml_axis,:].mean()





  
  

  1052
  
    middle_lat_cart = lat_cart_locs[ml_axis,:].mean()





  
  

  1053
  





  
  

  1054
  
    #get location of tibia to get centroid of tibial plateau





  
  

  1055
  
    tib_locs = np.asarray(np.where(seg_array == tib_label))





  
  

  1056
  
    middle_tib = tib_locs[ml_axis, :].mean()





  
  

  1057
  
    center_tibia_slice = int(middle_tib)





  
  

  1058
  





  
  

  1059
  
    # infer the direction(s) for medial/lateral





  
  

  1060
  
    med_direction = np.sign(middle_med_cart - middle_tib)





  
  

  1061
  
    lat_direction = np.sign(middle_lat_cart - middle_tib)





  
  

  1062
  
    if med_direction == lat_direction:





  
  

  1063
  
        raise Exception(&#39;Middle of med and lat tibial cartilage on same side of centerline!&#39;)





  
  

  1064
  





  
  

  1065
  
    #create med/lat cartilage masks - binary for updating seg masks





  
  

  1066
  
    med_tib_cart_mask = np.zeros_like(seg_array)





  
  

  1067
  
    lat_tib_cart_mask = np.zeros_like(seg_array)





  
  

  1068
  





  
  

  1069
  
    if med_direction &gt; 0:





  
  

  1070
  
        med_tib_cart_mask[center_tibia_slice:,...] = 1





  
  

  1071
  
        lat_tib_cart_mask[:center_tibia_slice,...] = 1





  
  

  1072
  
    elif med_direction &lt; 0:





  
  

  1073
  
        med_tib_cart_mask[:center_tibia_slice,...] = 1





  
  

  1074
  
        lat_tib_cart_mask[center_tibia_slice:,...] = 1





  
  

  1075
  





  
  

  1076
  
    # create new med/lat cartilage arrays 





  
  

  1077
  
    new_med_cart_array = array_tib_cart * med_tib_cart_mask





  
  

  1078
  
    new_lat_cart_array = array_tib_cart * lat_tib_cart_mask





  
  

  1079
  





  
  

  1080
  
    #make copy of original segmentation array &amp; update





  
  

  1081
  
    # med/lat tibial cartilage labels





  
  

  1082
  
    new_seg_array = seg_array.copy()





  
  

  1083
  
    new_seg_array[new_med_cart_array == 1] = med_tib_cart_label





  
  

  1084
  
    new_seg_array[new_lat_cart_array == 1] = lat_tib_cart_label





  
  

  1085
  
    





  
  

  1086
  
    return new_seg_array</code></pre>





  
  

  1087
  
</details>





  
  

  1088
  
</dd>





  
  

  1089
  
</dl>





  
  

  1090
  
</section>





  
  

  1091
  
<section>





  
  

  1092
  
</section>





  
  

  1093
  
</article>





  
  

  1094
  
<nav id="sidebar">





  
  

  1095
  
<h1>Index</h1>





  
  

  1096
  
<div class="toc">





  
  

  1097
  
<ul></ul>





  
  

  1098
  
</div>





  
  

  1099
  
<ul id="index">





  
  

  1100
  
<li><h3>Super-module</h3>





  
  

  1101
  
<ul>





  
  

  1102
  
<li><code><a title="pymskt.image" href="index.html">pymskt.image</a></code></li>





  
  

  1103
  
</ul>





  
  

  1104
  
</li>





  
  

  1105
  
<li><h3><a href="#header-functions">Functions</a></h3>





  
  

  1106
  
<ul class="">





  
  

  1107
  
<li><code><a title="pymskt.image.cartilage_processing.CofM" href="#pymskt.image.cartilage_processing.CofM">CofM</a></code></li>





  
  

  1108
  
<li><code><a title="pymskt.image.cartilage_processing.absolute_CofM" href="#pymskt.image.cartilage_processing.absolute_CofM">absolute_CofM</a></code></li>





  
  

  1109
  
<li><code><a title="pymskt.image.cartilage_processing.findNotch" href="#pymskt.image.cartilage_processing.findNotch">findNotch</a></code></li>





  
  

  1110
  
<li><code><a title="pymskt.image.cartilage_processing.getAnteriorOfWeightBearing" href="#pymskt.image.cartilage_processing.getAnteriorOfWeightBearing">getAnteriorOfWeightBearing</a></code></li>





  
  

  1111
  
<li><code><a title="pymskt.image.cartilage_processing.getCartilageSubRegions" href="#pymskt.image.cartilage_processing.getCartilageSubRegions">getCartilageSubRegions</a></code></li>





  
  

  1112
  
<li><code><a title="pymskt.image.cartilage_processing.get_knee_segmentation_with_femur_subregions" href="#pymskt.image.cartilage_processing.get_knee_segmentation_with_femur_subregions">get_knee_segmentation_with_femur_subregions</a></code></li>





  
  

  1113
  
<li><code><a title="pymskt.image.cartilage_processing.get_y_CofM" href="#pymskt.image.cartilage_processing.get_y_CofM">get_y_CofM</a></code></li>





  
  

  1114
  
<li><code><a title="pymskt.image.cartilage_processing.verify_and_correct_med_lat_tib_cart" href="#pymskt.image.cartilage_processing.verify_and_correct_med_lat_tib_cart">verify_and_correct_med_lat_tib_cart</a></code></li>





  
  

  1115
  
</ul>





  
  

  1116
  
</li>





  
  

  1117
  
</ul>





  
  

  1118
  
</nav>





  
  

  1119
  
</main>





  
  

  1120
  
<footer id="footer">





  
  

  1121
  
<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.10.0</a>.</p>





  
  

  1122
  
</footer>





  
  

  1123
  
</body>





  
  

  1124
  
</html>