Paper title — replace with your first publication

P. C. Kalaiselvi, Co-author, Co-author

arXiv preprint · In preparationPlaceholder

Paper title — replace with a workshop or conference paper

Co-author, P. C. Kalaiselvi, Co-author

Workshop / Conference · Under reviewPlaceholder

Paper title — replace with a technical report

P. C. Kalaiselvi

Technical report · DraftPlaceholder

Flow-state modeling for a Cognitive Performance OS

Applied research on flow-state modeling and closed-loop neurostimulation, grounded in Csikszentmihalyi's Flow Theory. Statistical models for skill decay, exponential moving average mastery estimation, and flow-channel classification from behavioral data.

Hyle Global · Cognition, Statistical Modeling

Contextual memory for an AI wearable

Kenesis Vision: a lightweight AI smartglass for real-time scene understanding, navigation and memory recall, with a hybrid edge-cloud engine for low-connectivity use, extended into a modular API layer for reasoning and episodic memory.

Kenesis Labs · Computer Vision, Episodic Memory

Skeletal bone-age estimation from pediatric hand radiographs

An end-to-end CNN pipeline for bone-age estimation, deployed with MONAI Label and 3D Slicer and validated on-site at Egmore Children's Hospital against live clinical X-rays and radiologist assessments.

OsteQuant · IIT Madras Research Park · Medical Imaging, Deep Learning

Automated bus scheduling and route optimization

Adaptive Large Neighbourhood Search for driver duty scheduling, LSTM models for passenger flow and arrival-time prediction, and GIS-based route optimization for Delhi Transport Corporation.

Smart India Hackathon 2024 · Optimization, Forecasting

Plant health from bioelectrical signals

Classifying plant bioelectrical signals into healthy or stressed and hydrated or dehydrated states, using a Bio Amp ExG Pill, Arduino, BrainBay and Azure Machine Learning.

TerraIntel · Microsoft Imagine Cup 2024 · Biosignals, Classification

Attention Is All You Need

Vaswani et al. · NeurIPS

Deep Residual Learning for Image Recognition

He, Zhang, Ren, Sun · CVPR

U-Net: Convolutional Networks for Biomedical Image Segmentation

Ronneberger, Fischer, Brox · MICCAI

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Dosovitskiy et al. · ICLR

Learning Transferable Visual Models From Natural Language Supervision

Radford et al. · ICML

Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

Lewis et al. · NeurIPS

MONAI: An open-source framework for deep learning in healthcare

Cardoso et al. · arXiv