Welcome! I'm Suhrud! After finishing medical school and during my clinical internship, I had the opportunity to work at two of India’s premier institutions – the Indian Institute of Science (IISc), Bangalore, and the Indian Institute of Technology (IIT), Kharagpur – focusing on clinical research and artificial intelligence in healthcare.
The first artificial neural network was proposed by Dr. Warren McCulloch in 1943, a neuropsychologist by training who worked at MIT with Walter Pitts. He was so inspired by the brain's microscopic architecture that they modeled this artificial neural network based on a frog's visual system. I, too, have been similarly awestruck by the brain's structural connectivity with its physiological resilience and adaptability since my first year of medical school, and believe that the secrets to building the next generation of Artificial General Intelligence (AGI) can be found here. I am passionate about neurology and am eager to pursue my residency while working to integrate explainable artificial intelligence (XAI) models into clinical workflows. During my medical internship, I worked in the general neurology service at a large 1100-bed tertiary care teaching hospital in Pune, India. Additionally, I rotated in the neurology ward service at Yale New Haven Hospital, the movement disorders clinic at Mayo Clinic, and vascular and interventional neurology at the University of Miami Jackson Memorial Hospital.
My love for mathematics in high school sublimated into a passion for statistics and probability theory during med school. I have a YouTube channel where I upload simplified follow-along tutorials for medical students and residents, demonstrating how to practically perform statistical tests using open-source software. By demystifying complex statistical models and interpreting research data, I strive to make biostatistics more accessible and relevant for healthcare professionals. My current research interests include interpretable predictive modeling using machine and deep learning algorithms. As an editorial board member for Nature’s 'Scientific Reports,' I have the distinct privilege of using and honing my expertise in research methodology to critically evaluate studies across a broad spectrum of neurology subspecialties, focusing on neuroinformatics and predictive models.
I am interested in integrating computer vision with trustable generative AI to not only expedite and improve neurodiagnosis but also contribute to personalized medicine. Unlike surgical specialties, where procedures are relatively personalized, medicine specialties such as neurology have yet to explore treatment options, optimal doses, and potential adverse effects specific to patients. To this end, given enough data, predictive AI models can help to achieve this and go even further, improving routine healthcare outcomes and patient satisfaction. Having experience training interpretable deep learning models to make novel radiomic and time-series predictions in neuro-oncology, neurocritical care, behavioral neurology, and movement disorders, my interest lies in furthering this pursuit of establishing personalized medicine. I also have experience developing, patenting, and deploying user-friendly software that allows clinicians to input their imaging, sensor, and other relevant data to get robust predictions. I have been able to pursue this research in resource-limited settings using grants I have received for two consecutive years from the Indian Council of Medical Research. I have also been fortunate to receive two prestigious scholarships from Tata Trusts for excellence in Medicine and Healthcare during my medical school.
ReQuir Statistics Solutions is a statistical consultancy firm that I co-founded with 2 of my friends, which was created with the idea to bridge the gap between the needs of busy healthcare professionals for high-quality analytics with time sensitivity and mutual understanding. As the company evolved, so did our objectives, providing tailored solutions to larger hospitals and health AI firms and helping them make data-driven decisions.
I have shared some programming codes for real-world problems in my GitHub repositories. I have also included tutorial codes for the intermediate programmer using simulated datasets in R, SAS, Python, and Julia programming languages.
I love to play all kinds of sports, but the ones closest to my heart are calisthenics, stair-climbing, and table tennis. In my free time, I also play the guitar, love to experiment in the kitchen, am a PADI-certified open-water scuba diver, and admire the color- and strokeplay by the 19th and 20th-century impressionists!
Please feel free to get in touch with me!