Before I joined IIIT Hyderabad, I was a Research Fellow at iHub Data, where I worked with Prof. Bapi Raju S and Prof. U Deva Priyakumar. Prior to that, I was a Data Engineering intern at BlackRock where I worked on developing a real-time ML pipeline to predict trade settlement failure. I (used to) maintain an open-source project AutoSub which is a tool to automatically generate subtitles for any video file (locally). I graduated with a B.Tech. in Computer Science and Engineering from Manipal Institute of Technology, Manipal, India in 2017. For more details, refer to my resume or drop me an email.
Self-Supervision and Weak Supervision for Accurate and
Interpretable Chest X-Ray Classification Models
, Akshaya Karthikeyan, Shanmukh Alle, Maitreya Maity, U. Deva Priyakumar
IJCNN'23
Current state of artificial intelligence applications in ophthalmology
and their potential to influence clinical practice
Dasharathraj K Shetty, , Swapna Shanbhag, B.M Zeeshan Hameed, Nithesh Naik
Best Poster Presentation Award @ USICON'21 | Cogent Engineering'21
Application of DCNN
in prediction of stone location, skin to stone distance and
composition in renal lithiasis
B.M Zeeshan Hameed, Bhaskar Somani, Nithesh Naik, , Milap Shah, Sourabh Reddy
European Urology'21
International Institute of Information Technology
Research Fellow, Healthcare & Artificial Intelligence (HAI)
- Worked on improving the interpretability of classification models for disease diagnostics using self-supervision within the purview of Chest X-rays
- Achieved downstream multi-label classification scores comparable to current supervised SoTA and superior performance in terms of Grad-CAM localization
- Ranked 3rd in the X-ray Projectomic Reconstruction Challenge hosted by Harvard Medical School. Developed an Attention-inspired U-Net to predict axon trajectories in volumetric XNH images. Presented at ISBI'23
Data Engineering Intern
- Developed a real-time ML pipeline to predict trade settlement failure using large, imbalanced, and distributed datasets
- Worked with financial and regulatory teams to perform EDA and pre-process datasets
- Experimented with models like XGBoost and SVM and achieved a fail-capture rate of 84%
Machine Learning Intern
- Trained and integrated an Automatic Speech Recognition model for a video-conferencing platform
- Improved word error rate by 30% over benchmark with custom fine-tuned models for better generalization to the Indian Accent using Indic Speech data
- Implemented RESTful APIs as Linux systemd processes, improving TAT by 50%
International Institute of Information Technology, Hyderabad
Master of Science by Research in CSE | CGPA: 9.2 | July'22 – Present
Manipal Institute of Technology, Manipal
Bachelor of Technology in CSE | July'17 – July'21
AutoSub
- Developed a CLI application to generate subtitles for video files on-device automatically
- Implemented MFCC features to segment audio on non-speech segments and perform speech recognition
- Improved performance using an external scorer (language model) and added support for GPU-based inference
from scratch
minimal implementations from scratch of the following:
- baby86: a minimal x86 "bootloader" to print stuff on screen
- nn: dense neural network with multiple layers and activation functions
- cnn: NumPy-only CNN with Conv and MaxPool layers
- torch: not-so-minimal implementation of the torch API
Bioactivity Prediction
- Used regression models to predict biological activity (pIC50 values) of protein targets from ChEMBL database
- Calculated Lipinski and PaDEL descriptors using Acetylcholinesterase (AChE) as the target protein
- The best Decision Tree Regressor model achieved an R-squared value of 0.86
Augmented Random Search for Data Augmentation
- Improved AutoAugment by replacing the discrete search space with continuous space for augmentation policies
- Used Augmented Random Search method to improve performance and maintain diversities between sub-policies
Antenatal Care (iOS App)
- Created an iOS application using Swift and XCode to provide antenatal care for rural populations
- Implemented NFC to store electronic health records like test results, scans, prescription details on an NFC-enabled card
- Used Firebase as a back-end database for storage and retrieval of patient details