I work at the intersection of Machine Learning, Human-Computer Interaction, and Healthcare, developing AI systems that are both clinically meaningful and privacy-preserving.

Recognition & Fellowships

  • NSF NRT LINDIV Graduate Fellowship (2022-2023)
  • Google CS Research Mentorship Program (2022)
  • Google Developers ML Bootcamp (2022)
  • Penn State Graduate Student Travel Grant ($1,000)
  • Best Poster Awards: AI Week 2025 (3rd place), MASC-SLL 2025 (1st place)

Academic Journey

I have the same information in two sections here:

a) Longer Version: TS; WM (Too Short; Want More) and
b) Shorter Version: TL; DR (Too Long; Didn’t Read).

Pick yours!

TL;DR

My Google Scholar profile can be viewed here.

TS; WM

I graduated from PES University in Bangalore, India, with a B.Tech in Electronics & Communication Engineering, specializing in Signal Processing, and a minor in Computer Science. My thesis focused on using wireless sensor networks to prevent train derailments in India, and I was fortunate to be advised by Gp Capt. (Retd.) Suresh Padmanabhan.

Our work caught the attention of Shri. Piyush Goyal, the current Hon’ble Minister for Railways in India, who mentioned our approach (Building smarter (and safer) Indian Railways) in his plans for improving railway safety throughout the country. We were also able to present our work at the IEEE IC4 2018 conference in Thiruvananthapuram, India.

I was lucky to have the opportunity to work at the Indian Institute of Science in Bangalore on a project sponsored by the Department of Science and Technology (DST, Govt. of India). The project aimed to use speech signal processing to identify neurological conditions such as Amyotrophic Lateral Sclerosis and Parkinson’s disease. As part of my assistantship, I worked with patients at the National Institute of Mental Health and Neurosciences (NIMHANS), collecting and analyzing their speech samples. My advisor was Dr. Prasanta Kumar Ghosh, and I have been able to publish our findings at conferences such as INTERSPEECH 2019, ICASSP 2020, and SPCOM 2020 (View the paper presentation video here).

At Penn State, I have been fortunate to collaborate with researchers from different departments, including the School of EECS and the College of IST. For my Master’s thesis, I worked with Dr. Saeed Abdullah to explore how federated learning can be used to preserve privacy while assessing depression. I was honored to have Dr. David Miller and Dr. Jing Yang serve on my thesis committee, and our paper on this work was accepted at ICASSP 2022!

Recent Research (Ph.D. Thesis)

My doctoral research focused on Generative Human-Centered AI for Mental Health, developing novel approaches for PTSD therapy support:

  • Synthetic Therapy Data: Created “Thousand Voices of Trauma” - 3,000 synthetic PTSD therapy conversations with clinical validation
  • Empathetic AI: Developed TIDE dataset and fine-tuned small language models for trauma-informed, empathetic dialogue support
  • Automated Clinical Assessment: Built audio-language models for automatic temporal localization of therapy fidelity elements
  • Privacy-Preserving Methods: Extended federated learning, and differential privacy approaches for sensitive healthcare applications
  • Healthcare Applications: Continued work on cuff-less blood pressure monitoring and speech-based health assessment

You can visit the Talks section to view some of my presentations.

Fun Stuff

My research Co-authorship distances are as follows (in no particular order):

  • Dijkstra number of 5.
  • Erdős number of 5.
  • In the deep learning community: Andrew Ng (3), Yann LeCun (3), Geoff Hinton (4), and Yoshua Bengio (4).

Recent Publications & Patents

Patents

  1. [Published] Suhas BN, Y. M. Saidutta, R. S. Srinivasa, J. Cho, C.H. Lee, C. Yang, Y. Shen, and H. Jin. “Method for end-to-end cuff-less blood pressure monitoring using ECG and PPG signals.” U.S. Patent Application 18/883,902, filed March 20, 2025. (Google Patents Link)

Publications

2025-2026 (Submitted/In Press)

  1. [Submitted to ICASSP 2026] Suhas BN, J. Alaparthi, S. Abdullah., D. Mattioli, R. Arriaga, C. Wiese, and A. Sherrill. “Fine-Tuning Large Audio-Language Models with LoRA for Precise Temporal Localization of Prolonged Exposure Therapy Elements” (arXiv:2504.14821)

  2. [Submitted to IEEE JSTSP] Suhas BN, T. Bhattacharjee, B. K. Yamini, N. Atchayaram, R. Yadav, D. Gope, and PK Ghosh. “Contrastive Audio-Video Pretraining for Classification of Amyotrophic Lateral Sclerosis, Parkinson’s Disease, and Healthy Controls Using Smartphone Recordings” (arXiv:2504.18903)

  3. [Submitted to NeurIPS 2025] Suhas BN, A. Sherrill, R. Arriaga, C. Wiese, and S. Abdullah. “Thousand Voices of Trauma: A Large-Scale Synthetic Dataset for Modeling Prolonged Exposure Therapy Conversations” (arXiv:2504.13955)

  4. [Accepted at EMNLP 2025] Suhas BN, Y. Mahajan, S. Abdullah., D. Mattioli, R. Arriaga, C. Wiese, and A. Sherrill. “The Pursuit of Empathy: Evaluating Small Language Models for PTSD Dialogue Support” (arXiv:2505.15065)

  5. [Accepted at EMNLP 2025] Suhas BN, S. Abdullah., D. Mattioli, R. Arriaga, C. Wiese, and A. Sherrill. “How Real Are Synthetic Therapy Conversations? Evaluating Fidelity in Prolonged Exposure Dialogues” (arXiv:2504.21800)

2025

  1. [INTERSPEECH 2025] Suhas BN, JP. Cohen, HC. Shing, JC. Moriarty, L. Xu, M. Strong, J. Burnsky, J. Ofor, J.R. Mason, S. Chen, C. Shivade, S. Srinivasan. “Fact-Controlled Diagnosis of Hallucinations in Medical Text Summarization” (Paper) (arXiv:2504.16789)

2024

  1. [INTERSPEECH 2024] Suhas BN, A. Rebar, S. Abdullah. “Speaking of Health: Leveraging Large Language Models to assess Exercise Motivation and Behavior of Rehabilitation Patients” (Paper)

  2. [ICASSP 2024] Suhas BN, R.S. Srinivasa, Y.M. Saidatta, J. Cho, C.H. Lee, C. Yang, Y. Shen, H. Jin. “End-to-End Personalized Cuff-less Blood Pressure Monitoring using ECG and PPG Signals” (Paper)

2023

  1. [INTERSPEECH 2023] Suhas BN, S. Rajtmajer, S. Abdullah. “Privacy-Preserving Dementia Classification with Differential Privacy: An Exploration of the Privacy-Accuracy Tradeoff in Speech Signal Data” (Paper)

  2. [Springer Nature 2023] HJ. Han, Suhas BN, L. Qiu, S. Abdullah. “Automatic classification of dementia using text and speech data.” In Multimodal AI in Healthcare, pp. 399-407. (Chapter)

2022

  1. [ICASSP 2022] Suhas BN, S. Abdullah. “Privacy Sensitive Speech Analysis using Federated Learning to assess Depression.” (Paper)

2020

  1. [CogMI 2020] Suhas BN. “Automatic bird sound detection in long range field recordings using Wavelets & Mel filter bank features”. (Paper) (Code)

  2. [SPCOM 2020] Suhas BN, J. Mallela, A. Illa, B. K. Yamini, N. Atchayaram, R. Yadav, D. Gope, and PK Ghosh. “Speech task-based automatic classification of ALS and Parkinson’s Disease and their severity using log Mel spectrograms.” (Paper) (Code)

  3. [ICASSP 2020] Mallela, J, A Illa, Suhas BN, S. Udupa, Y. Belur, N. Atchayaram, R. Yadav, P. Reddy, D. Gope, and PK Ghosh. “Voice-based classification of patients with Amyotrophic Lateral Sclerosis, Parkinson’s Disease and Healthy Controls with CNN-LSTM using transfer learning.” (Paper)

2019

  1. [INTERSPEECH 2019] Suhas BN, D. Patel, NR Koluguri, Y. Belur, P. Reddy, A. Nalini, R. Yadav, D. Gope, and PK Ghosh. “Comparison of Speech Tasks and Recording Devices for Voice-Based Automatic Classification of Healthy Subjects and Patients with Amyotrophic Lateral Sclerosis.” (Paper)

2018

  1. [IC4 2018] Suhas BN, S. Bhagavat, V. Vimalanand, and P. Suresh. “Wireless Sensor Networks Based Monitoring of Railway Tracks.” (Paper)