Publications
Pathologists must assess bladder layers from tissue biopsies which are subject to variability among pathologists. We present a deep learning automated image analysis algorithm that recognizes different layers of bladder wall.
Read more here
Contributors: MKK Niazi, T Tavolara, V Arole, C Lee, AV Parwani, MN Gurcan
Pathologists need better tools to more accurately diagnose bladder cancer. To do this different layers of the bladder must be recognized which this abstract addresses.
Read more here
Contributors: MKK Niazi, T Tavolara, V Arole, C Lee, AV Parwani, MN Gurcan
Neutrophils are white blood cells that attack bacteria and potentially host tissue. Detection of neutrophils are necessary in the disease tuberculosis as neutrophils are biomarkers of lung damage. Utilizing transfer learning the method was able to identify neutrophil clusters at 93% accuracy.
Read more here
Contributors: MKK Niazi, G Beamer, MN Gurcan
Read more here
Contributors: MKK Niazi, G Beamer, MN Gurcan
Automated staging of T1 bladder cancer using digital pathologic H&E images: a deep learning approach
The paper discusses the need for accurately gauging tumor cell intrusion into Lamina Propria in an effort to substage bladder cancer. It explains how transfer learning in conjunction with Convolutional Neural Networks can be used to accurately identify different bladder layers and then compute the distance between tumor nuclei and Lamina Propria.
Read more here
Contributors: Muhammad Khalid Khan Niazi, Thomas Tavolara, Vidya Arole, Anil Parwani, Cheryl Lee, Metin Gurcan
Read more here
Contributors: Muhammad Khalid Khan Niazi, Thomas Tavolara, Vidya Arole, Anil Parwani, Cheryl Lee, Metin Gurcan
Bladder layer identification is a necessity for proper bladder cancer diagnosis. This article presents an automated Bladder Image Analysis System (BLIAS) that can differentiate layers of the bladder.
Read more here
Contributors: M Khalid Khan Niazi, Thomas E Tavolara, Vidya Arole, Anil V Parwani, Cheryl Lee, Metin N Gurcan
Read more here
Contributors: M Khalid Khan Niazi, Thomas E Tavolara, Vidya Arole, Anil V Parwani, Cheryl Lee, Metin N Gurcan
Tumor budding is a recently recognized, independent prognostic factor in colorectal cancer, but lacks a standardized assessment methodology. Although staining with pan-cytokeratin has been shown to mitigate the issue of lack of reproducible, intra-observer agreement, usage of this antibody remains expensive and is limited in clinical practice. We propose an automated image analysis framework to take advantage of the visual superiority of pan-cytokeratin and the routine use of H&E to detect and quantify tumor budding. Our framework has demonstrated promising ability to identify tumor regions of colorectal slides – 92.0% accuracy, 94.5% sensitivity, and 85% specificity – across four independent datasets. Read more here
Contributors: Thomas E. Tavolara, M. Khalid Khan Niazi, Wei Chen, Wendy Frankel, Metin N. Gurcan
The aim of this study is to develop a deep learning based software called, DeepFocus, which can automatically detect and segment blurry areas in digital whole slide images to address these problems. DeepFocus is built on TensorFlow, an open source library that exploits data flow graphs for efficient numerical computation. DeepFocus was trained by using 16 different H&E and IHC-stained slides that were systematically scanned on nine different focal planes, generating 216,000 samples with varying amounts of blurriness. DeepFocus has the potential to be integrated with whole slide scanners to automatically re-scan problematic areas, hence improving the overall image quality for pathologists and image analysis algorithms.
Contributors: Caglar Senaras, M. Khalid Khan Niazi, Gerard Lozanski, and Metin Nafi Gurcan
Identifying an abnormal eardrum can be a difficult task for untrained or lesser experienced individuals. We present a deep learning method that automatically assess an eardrum as normal or abnormal utilizing transfer learning and the development of a neural network.
Contributors: Caglar Senaras, Aaron C. Moberly, Theodoros Teknos, Garth Essig, Charles Elmaraghy, Nazhat Taj-Schaal, Lianbo Yua, Metin N. Gurcan
In modern clinical practice, digital pathology has a crucial role and is increasingly a technological requirement in the scientific laboratory environment. The advent of whole-slide imaging, availability of faster networks, and cheaper storage solutions has made it easier for pathologists to manage digital slide images and share them for clinical use. In parallel, unprecedented advances in machine learning have enabled the synergy of artificial intelligence and digital pathology, which offers image-based diagnosis possibilities that were once limited only to radiology and cardiology. Integration of digital slides into the pathology workflow, advanced algorithms, and computer-aided diagnostic techniques extend the frontiers of the pathologist's view beyond a microscopic slide and enable true utilization and integration of knowledge that is beyond human limits and boundaries, and we believe there is clear potential for artificial intelligence breakthroughs in the pathology setting. In this Review, we discuss advancements in digital slide-based image diagnosis for cancer along with some challenges and opportunities for artificial intelligence in digital pathology.
Contributors: M. Khalid Khan Niazi, Anil V. Parwani, Metin Nafi Gurcan
The morphological features that pathologists use to differentiate neoplasms from normal tissue are nonspecific to tissue type. For example, if given a Ki67 stained biopsy of neuroendocrine or breast tumor, a pathologist would be able to correctly identify morphologically abnormal cells in both samples but may struggle to identify the origin of both samples. This is also true for other pathological malignancies such as carcinomas, sarcomas, and leukemia. This implies that computer algorithms trained to recognize tumor from one site should be able to identify tumor from other sites with similar tumor subtypes. Here, we present the results of an experiment that supports this hypothesis. We train a deep learning system to distinguish tumor from non-tumor regions in Ki67 stained neuroendocrine tumor digital slides. Then, we test the same, unmodified, deep learning model to distinguish breast cancer from non-cancer regions. When applied to a sample of 96 high power fields, our system achieved a cumulative pixel-wise accuracy of 86% across these high-power fields. To our knowledge, our results are the first to formally demonstrate generalized segmentation of tumors from different sites of origin through image analysis. This paradigm has the potential to help with the design of tumor identification algorithms as well as the composition of the datasets they draw from.
Contributors: Muhammad Khalid Khan Niazi; Thomas E. Tavolara; Caglar Senaras; Gary Tozbikian M.D.; Douglas J. Hartman M.D.; Vidya Arole M.D.; Liron Pantanowitz M.D.; Metin N. Gurcan M.D.
Verifying the accuracy of nuclei detection algorithms can be difficult due to the requirement of acquiring manually annotated ground truth from pathologists and their inherent variability. This paper proposes a method for creating digital immunohistochemistry (IHC) phantoms that can be used to evaluate computer algorithms for enumeration of IHC positive cells.
Contributors: Muhammad Khalid Khan Niazi, FS Abas, C Senaras, M Pennell, B Sahiner, Weijie Chen, John Opfer, R Hasserjian, Abne Louissaint Jr., Arwa Shana'ah, Gerard Lozanski, MN Gurcan
Typically, immunohistochemical staining interpretation is rendered by a trained pathologist using a manual method, which consists of counting each positively- and negatively-stained cell under a microscope. The manual interpretation results in poor reproducibility. To address this issue, we proposed a novel method to create artificial datasets with the known ground truth allowing us to analyze recall, precision, accuracy, and intra- and inter-observer variability in a systematic manner, and enabling us to compare different computer analysis approaches.
Contributors: Caglar Senaras, Muhammad Khalid Khan Niazi, Berkman Sahiner, Michael P. Pennell, Gary Tozbikian, Gerard Lozanski, Metin N. Gurcan
We proposed a pathological image compression framework to address the needs of Big Data image analysis in digital pathology. Big Data image analytics require analysis of large databases of high-resolution images using distributed storage and computing resources along with transmission of large amounts of data between the storage and computing nodes that can create a major processing bottleneck. The proposed image compression framework is based on the JPEG2000 Interactive Protocol and aims to minimize the amount of data transfer between the storage and computing nodes as well as to considerably reduce the computational demands of the decompression engine. The proposed framework was integrated into hotspot detection from images of breast biopsies, yielding considerable reduction of data and computing requirements.
Contributors: M. Khalid Khan Niazi, Y Lin, F Liu, A Ashok, MW Marcellin, Tozbikian G, MN Gurcan, A Bilgin
Follicular Lymphoma (FL) is the second most common subtype of lymphoma in the Western World. It is a low-grade lymphoma arising from Germinal Centre (GC) B cells. The neoplasm predominantly consists of back-to-back arrangement of nodules or follicles of transformed GC B cells with the replacement of lymph node architecture and loss of normal cortex and medullary differentiation, which is preserved in non-neoplastic or reactive lymph node. There is a growing interest in studying different cell subsets inside and on the periphery of the follicles to direct curative therapies and minimize treatment-related complications. To facilitate this analysis, we develop an automated method for follicle detection from images of CD8 stained histopathological slides. The proposed method is trained on eight whole digital slides. The method is inspired by U-net to segment follicles from the whole slide images. The results on an independent dataset resulted in an average Dice similarity coefficient of 85.6% when compared to an expert pathologist’s annotations. We expect that the method will play a considerable role for comparing the ratios of different subsets of cells inside and at the periphery of the follicles.
Contributors: C. Senaras; M. K. K. Niazi; V. Arole; W. Chen; B. Sahiner; A. Shana’ah; A. Louissaint; R. P. Hasserjian; G. Lozanski; M. N. Gurcan