Dr.Raj Subramanian, PhD, PMP

Histological grading quantifies the tumor architecture and the cytology deviation of breast cancer against normal tissue. Nottingham Grading System grades the breast cancer classification and allot tumor scores. Mitotic detection is one of the major component in the Nottingham Grading System. Using a conventional microscope is time-consuming, semi-quantitative and have limited histological parameters. Digital scanners scan the tissue slice into high-resolution virtual images called whole slide… Show more

Histological grading quantifies the tumor architecture and the cytology deviation of breast cancer against normal tissue. Nottingham Grading System grades the breast cancer classification and allot tumor scores. Mitotic detection is one of the major component in the Nottingham Grading System. Using a conventional microscope is time-consuming, semi-quantitative and have limited histological parameters. Digital scanners scan the tissue slice into high-resolution virtual images called whole slide images. Deep learning models on whole slide images provide a fast and accurate quantitative diagnosis. This paper proposes two deep learning models namely Faster R-CNN and YOLOv5 to detect mitosis on WSI. The proposed Deep Learning models uses 56258 annotated tiles for training/testing and provide F1 score as 84%. The proposed model uses a web-based imaging analysis and diagnosis platform called CADD4MBC for image uploading, Annotation and visualization. This paper proposes an end-to-end web based Deep Learning detection for Breast Cancer Mitosis. Show less

Epithelium tissue covers and lines all internal organs of the body. Breast cancer carcinomas arise from
the epithelial cells of the breast. The epithelial component lines tubules, ducts, etc. and segmenting out
this region from other tissues is important to detect breast cancer. This paper proposes application of deep learning technique U-Net, to segment epithelium tissue automatically from a whole slide image (WSI) image. It also implements U-Net, an image segmentation algorithm to… Show more

Epithelium tissue covers and lines all internal organs of the body. Breast cancer carcinomas arise from
the epithelial cells of the breast. The epithelial component lines tubules, ducts, etc. and segmenting out
this region from other tissues is important to detect breast cancer. This paper proposes application of deep learning technique U-Net, to segment epithelium tissue automatically from a whole slide image (WSI) image. It also implements U-Net, an image segmentation algorithm to automatically learn the features of epithlium components from the experts-annotated WSI image dataset and tested the results. Various image processing techniques such as thresholding are applied to improve the quality of dataset before and after the training of images by U-Net. The performance of the U-Net is measured by statistic parameters Sørensen–Dice coefficient and F1 score. The automatic system generated an epithelium segmentation of accuracy of 0.932. Show less

The evaluation of lymph nodes’ metastasis is an important component of Tumor, Node, Metastasis
(TNM) breast cancer staging system for better clinical management and treatment. Assessing lymph node metastasis through histologic examination is the most accurate method. This paper proposes significantly advanced and faster image classification Convolutional Neural Network (CNN) model called Densenet-161 for lymph node metastasis. This paper uses pre-processing technique called image… Show more

The evaluation of lymph nodes’ metastasis is an important component of Tumor, Node, Metastasis
(TNM) breast cancer staging system for better clinical management and treatment. Assessing lymph node metastasis through histologic examination is the most accurate method. This paper proposes significantly advanced and faster image classification Convolutional Neural Network (CNN) model called Densenet-161 for lymph node metastasis. This paper uses pre-processing technique called image thresholding to improve the contrast intensities of the SLN images, which improves the performance of DenseNet. The experimental PCam dataset contains 327,680 patches extracted from 400 Haemotoxylin and Eosin (H&E) stained WSIs of breast cancer with sentinel lymph node sections. The proposed system has generated 94% accuracy for lymph node metastasis classification Show less

DNA-Drug design using identified FCPs in Human Genome Database. Implemented using MongoDB and Java

The identification of short length of FCPs of length between 6 to 18 is needed in the following genomic sequence analysis functions and genetic pattern discovery tools development. (1). Repetitive DNA in Human Genome (2). Cis-regulatory modules (CRM) (3). Sequence Motif (4).Transcription Factor (5).Promoter of gene (6).Single Nucleotide Polymorphism (SNP).

Positions based Big Data algorithm to Identify n_length Frequent Contiguous Patterns (FCP) in a Large Human Genome Sequence Database, implemented using Java and MySQL

Transaction sequences in market-basket analysis have large set of alphabets with small length, whereas bio-sequences have small set of alphabets of long length with gap. There is the difference in pattern finding algorithms of these two sequences. The chances of repeatedly occurring small patterns are high in bio-sequences than in the transaction sequences. These repeatedly occurring small patterns are called as Frequent Contiguous Patterns (FCP). The challenging task in pattern finding of… Show more

Transaction sequences in market-basket analysis have large set of alphabets with small length, whereas bio-sequences have small set of alphabets of long length with gap. There is the difference in pattern finding algorithms of these two sequences. The chances of repeatedly occurring small patterns are high in bio-sequences than in the transaction sequences. These repeatedly occurring small patterns are called as Frequent Contiguous Patterns (FCP). The challenging task in pattern finding of bio-sequences is to find FCP. FCP gives clues for genetic discovery, functional analysis and also helps to assemble a whole genome of species. Most of the existing FCP algorithms are all based on Apriori method. They require repeated scanning of the database and large number of intermediate tables to produce the results. So, these algorithms require large space and high computational time. In this paper, we are analyzing few of the currently available FCP algorithms with their advantages and disadvantages. Show less

Proceedings of IEEE International Conference on Intelligent Computing Applications held at Bharathiar University on 6th, 7th March, 2014.

Print ISBN: 978-1-4673-2756-5, INSPEC Accession Number: 13291636,
Digital Object Identifier : 10.1109/ICRCC.2012.6450563

Identifying Frequent Contiguous Patterns (FCP) means to find the frequently repeated contiguous pattern in Sequence Database (SDB). This Algorithm helps in many SDB applications such as to extract Motif or Regulatory Regions in Genomic SDB, to find repeated requiring items in the Inventory SDB and to find stock trading patterns in Stock Trading SDB, etc.

I Co-authored this publication

Quantitative Financial Analysis field leverages computer technologies to build models from large amounts of data to extract insight. In today’s networked world, the amount of data available to build and refine models has been increasing exponentially. Often times the difference between seizing an opportunity and missing one is the latency involved in processing this data. The need to do complex processing with minimal latency over large volumes of data has led to the evolution of different data… Show more

Quantitative Financial Analysis field leverages computer technologies to build models from large amounts of data to extract insight. In today’s networked world, the amount of data available to build and refine models has been increasing exponentially. Often times the difference between seizing an opportunity and missing one is the latency involved in processing this data. The need to do complex processing with minimal latency over large volumes of data has led to the evolution of different data processing paradigms. Increasingly there is a need to develop an event-oriented application development paradigm to decrease the latency in processing large volumes of data. Recent technological advances have pushed the emergence of a new class of data-intensive applications that require continuous processing over sequences of transient data, called data streams, in near real-time.
This work presents a solid and powerful foundation for processing continuous queries over data streams. We (i) define a sound semantics for continuous sliding window queries, (ii) introduce a unique stream algebra implemented with efficient online algorithms, and (iii) sketch the adaptive runtime environment. Our work carries over and enhances findings from temporal databases to meet the challenging requirements of the data stream in Quantitative Financial Analysis Field. This will aid “Quants – Quantitative Financial Analysts” to develop Ideal solutions for High Data-intensive, Real-Time and Continuous Quantitative Finance Problems using Highly Efficient Data Processing Paradigm. Show less

A BigData Pattern decomposing algorithm to Identify Frequent Contiguous Patterns (FCP) from Large DNA Sequence Database using Permutations with Repetition (PR). This optimizes memory cache utilization. Implemented using Java and MySQL