-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathresearch.html
122 lines (103 loc) · 7.78 KB
/
research.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Document 1</title>
<div style="background-image: url('./images.jpg'); background-size: 100% 100%; width:1350px; height:800px";>
<div><style>
ul {
list-style-type: none;
margin: 0;
padding: 0;
overflow: hidden;
background-color: #333;
}
li {
float: left;
}
li a {
display: block;
color: white;
text-align: center;
padding: 14px 16px;
text-decoration: none;
}
li a:hover:not(.active) {
background-color: #111;
}
.active {
background-color: #78b8e0;
}
</style>
<ul>
<li style="float:right"><a href="#contact">Contact</a></li>
<li style="float:right"><a href="#news">News</a></li>
<li style="float:right"><a href="#about">Experience</a></li>
<li style="float:right" onclick="window.location.href='./index.html'";><a class="active" href="#home">Home</a></li>
</ul>
</div>
</head>
<body>
<div style=" width:55% ; float:left; border-color:rgba(64, 224, 208, 0)">
<h1 style="text-align: center;">Research Projects</h1>
<div style=" float:left; margin-bottom: 10px; border-style: solid; border-color: #78b8e0; padding-left: 4px; padding-right: 5px; padding-bottom:7px;">
<div style="width:20% ; float:left;margin-left: 1%; margin-top:7px; margin-bottom:0px;">
<img src='./gm.jpg' style=" margin-top: 0.1%; width:185px;height:185px;">
</div>
<div style="width:71.5% ; float:right; margin-right: 0.4%; margin-bottom:0px; " >
<h4 style="font-size:17px; padding: 3px;margin: 3px; margin-left:0px">
Automated Diagnosis of Ductal Carcinoma (Breast Cancer) from FNAC Images with Deep Learning </h4>
<h4 style="font-size:12.5px; text-align: justify; margin-bottom:0px;" >
This is a Computer-Aided Diagnosis project aimed at speeding up diagnosis and easing the Breast Cancer burden on diagnostic labs. Convolutional Neural Networks (CNNs) were used to detect microscopic cell clusters in whole slide images from Fine Needle Aspiration and classify them as benign (non-cancerous) or malignant (cancerous).
I labeled, annotated and augmented cytology images collected from two Pathology Labs. I then performed a comparative study on the efficiency of CNN classifiers (AlexNet, VGG16, 19, ResNet-50, Inception-V3) for cytology images and built a custom CNN classifier which achieved the highest accuracy (97%).
To minimize the time and effort needed for data preparation, I integrated YOLO for automatic Region Proposal and later used YOLO as a fully-independent pipeline for cell cluster detection and classification.
</h4>
</div>
</div>
<div style=" float:left; margin-bottom: 10px; border-style: solid; border-color: #78b8e0; padding-left: 4px; padding-right: 5px; padding-bottom:7px;">
<div style="width:20% ; float:left;margin-left: 1%; margin-top:7px; margin-bottom:0px;">
<img src='./ctscan.jpg' style=" margin-top: 0.1%; width:185px;height:185px;">
</div>
<div style="width:71.5% ; float:right; margin-right: 0.4%; margin-bottom:0px; " >
<h4 style="font-size:17px; padding: 3px;margin: 3px; ">
COVID-19 Diagnosis and Affected Lung Lesion Segmentation in CT-scans
</h4>
<h4 style="font-size:12.5px; text-align: justify; margin-bottom:0px;" >
This is a Computer-Aided Diagnosis project aimed at speeding diagnosis and easing the Breast Cancer burden on diagnostic labs. I have used Convolutional Neural Networks (CNNs) to detect microscopic cell clusters in whole slide images from Fine Needle Aspiration and classify them as benign (non-cancerous) or malignant (cancerous).
I have collected cytology images from two Pathology Laboratories in India and labeled, annotated and augmented them. I have done a comparative study of classifier CNNs, built a custom CNN classifier which achieved higher accuracy than standard architectures (97%), used YOLO for Region Proposal and then as a fully-independent pipeline for cell cluster detection and classification.
</h4>
</div>
</div>
<div style=" float:left; margin-bottom: 10px; border-style: solid; border-color: #78b8e0; padding-left: 4px; padding-right: 5px; padding-bottom:7px;">
<div style="width:20% ; float:left;margin-left: 1%; margin-top:7px; margin-bottom:0px;">
<img src='./pii.jpg' style=" margin-top: 0.1%; width:185px;height:185px;">
</div>
<div style="width:71.5% ; float:right; margin-right: 0.4%; margin-bottom:0px; " >
<h4 style="font-size:17px; padding: 3px;margin: 3px; ">
PII (Personally Identifiable Information) Identification and Redaction in Documents
</h4>
<h4 style="font-size:12.5px; text-align: justify; margin-bottom:0px;" >
This is a Computer-Aided Diagnosis project aimed at speeding diagnosis and easing the Breast Cancer burden on diagnostic labs. I have used Convolutional Neural Networks (CNNs) to detect microscopic cell clusters in whole slide images from Fine Needle Aspiration and classify them as benign (non-cancerous) or malignant (cancerous).
I have collected cytology images from two Pathology Laboratories in India and labeled, annotated and augmented them. I have done a comparative study of classifier CNNs, built a custom CNN classifier which achieved higher accuracy than standard architectures (97%), used YOLO for Region Proposal and then as a fully-independent pipeline for cell cluster detection and classification.
</h4>
</div>
</div>
</div>
<div style="margin: 50,50, 50,50; width:45% ; float:right; ">
<h1 style="text-align: center;">Research Papers</h1>
<div style=" float:left; margin-bottom: 20px; margin-left: 10%; margin-right: 7%;border-style: solid; border-color: #78b8e0; padding-left: 10px;">
<h4><i>"YOLO as a Region Proposal Network for Diagnosing Breast Cancer"</i>- Original Research paper accepted at Grace Hopper Celebration India 2020 (IEEE Conference ID: 50508) (17th-20th Feb 2021)</h4>
</div>
<h4 style="font-size: 2px;"> </h4>
<div style=" float:left; margin-bottom: 20px;margin-left: 10%; margin-right: 7%;border-style: solid; border-color: #78b8e0; padding-left: 10px; padding-right:10x;">
<h4><i>"Automated Diagnosis of Breast Cancer with RoI Detection using YOLO and Heuristics"</i>- Accepted at 17th ICDCIT (International Conference on Distributed Computing and Internet Technology) (7th-10th Jan 2021)
<a href='https://icdcit.ac.in/'>(here)</a></h4>
</div>
<div style=" float:left; margin-left: 10%; margin-right: 7%; border-style: solid; border-color: #78b8e0; padding-left: 10px;">
<h4>Under Revision:</h4>
<h4><i>"BFCNet: A CNN for Diagnosis of Ductal Carcinoma in Breast from Cytology Images"</i>- Pattern Analysis and Applications (Springer)</h4>
</div>
</div>
</body>
</html>