Predicting Respiratory Diseases from audio base input of coughing and sneezes.

The health economic models related to Asthma

Smart, portable spirometer with real-time lung function analysis and mobile app integration for accessible respiratory health monitoring.

Lung Sound Classification Models

AI-powered respiratory disease detection using deep learning on audio spectrograms | Published IEEE Research | Multi-class classification of 8 respiratory conditions including COPD, Asthma, and Pneumonia

MPH Capstone Data Management & Analysis

A Python-based dataset of high-quality respiratory sound recordings, annotated for machine learning tasks focused on detecting lung conditions like wheezes and crackles. It includes preprocessed audio, annotations, and subject metadata for research in respiratory health.

Interactive healthcare analytics dashboard built in Excel analyzing respiratory disease trends, treatment patterns, recovery outcomes, and state-wise variation using Pivot Tables, slicers, and KPI cards.

Code to accompany "Use of Patient-Reported Symptom Data in Clinical Decision Rules for Predicting Influenza in a Telemedicine Setting" by Billings et al, JABFM 2023

A full-stack deep learning application that analyzes lung or breath sounds to detect diseases like asthma, COPD, and pneumonia using spectrograms and a trained CNN model. Built using Flask, TensorFlow, and Librosa

Youreka 2425/ A cross-country public health analysis exploring associations between major depressive disorder prevalence and age-standardized respiratory infectious disease mortality using Pearson correlation, adjusted correlation, and GLM regression.

AI-powered respiratory disease detection system using lung sound analysis, signal processing, and machine learning.

Analysis of lung health measures from populations in Nepal, Peru and Uganda as part of the Global Excellence in COPD Outcomes-1 (GECo1) study. Funding: MRC (MR/P008984/1) & NIHR (303125). Paper: Alayadhi et al. 10.1183/13993003.01830-2025

Full stack AI diagnostic tool built with React, FastAPI, and TensorFlow. CNN trained on ICBHI 2017, async job queue with live terminal log, animated diagnosis progress with real time timer, built in sample recordings, PDF report export, and more. Dedicated UI for desktop, tablet, and mobile with multi theme support.

Android app for tracking symptoms in chronic respiratory disease patients. Built at BitsxLaMarató 2024.

Automated respiratory disease diagnosis from cough sounds using MFCC + Wavelet CNN. Detects cough with 97.4% accuracy and classifies COPD, Pneumonia, Bronchiectasis & more with 92% accuracy. Built with Python, Librosa & TensorFlow.