Project Description

Real-time biomarker analysis and prediction enable timely diagnoses and early personalized interventions for various disorders. Accurate physiology models can make human health digital twins (HHDTs) valuable for predicting treatment outcomes, reducing adverse reactions, and accelerating clinical trial regulatory processes. However, challenges include scarce longitudinal physiological data and the lack of physiology-informed AI models, particularly for neurodegenerative diseases (NDs), which lack effective preclinical phase indicators, progression trends, and treatment strategies. Current clinical methods for analyzing ND biomarkers are invasive, time-consuming, and complex. Wearable sensors provide extensive data collection from the human body but lack the necessary sensitivity and specificity for detecting ND biomarkers. While nanomaterials are highly sensitive to physiological signals, their use in wearable sensors is limited by intrinsic constraints. The AI-ARMADILLO framework will integrate multimodal wearable nanosensors, advanced AI algorithms, and collaborations with clinicians to develop HHDTs capable of simulating and predicting individuals' complex physiological and pathological states. The project aims to create nano-semiconductor-based wearable sensors to capture real-time physiological data streams with high sensitivity and specificity, facilitating physiology-informed deep learning for early disease diagnosis and continuous monitoring of disease progression. The developed platform will deliver clinically significant impacts across diverse user groups and activities.

Start Date

03/01/2025

Post Doc Qualifications

The candidate should have a PhD in material sciences, applied math, computer science, industrial engineering, electrical engineering, mechanical engineering, or related fields. Fluent programming in one of the following programming languages: Python/Julia. Familiar with PyTorch or TensorFlow. Research experience in artificial intelligence research, microfabrication, or wearable sensors is preferred.

Co-Advisors

Wenzhuo Wu
wu966@purdue.edu
Professor and University Faculty Scholar
School of Industrial Engineering
https://engineering.purdue.edu/wugroup


Guang Lin
guanglin@purdue.edu
Professor and University Faculty Scholar
Departments of Mathematics, Statistics & School of Mechanical Engineering
https://www.math.purdue.edu/~lin491/

Bibliography

1. Wang Y, Qiu G, Wang R, Huang S, Wang Q, Liu Y, Yuchen Du Y, Goddard W, Kim M, Xu X, Ye P, Wu W: Field-effect transistors made from solution-grown two-dimensional tellurene. Nature Electronics 1:228–236, 2018
2. Zhang, R.F., Jiang, J., Wu, W. Z.: Wearable chemical sensors based on 2D materials for healthcare applications. Nanoscale 15: 3079-3105, 2023
3. Haoyang Zheng, Jeffrey Petrella, P. Murali Doraiswamy, Guang Lin, Wenrui Hao, Data-driven causal model discovery and personalized prediction in Alzheimer’s disease, Nature NPJ Digital Medicine, 5, 137, 2022.
4. Ziqi Guo, Roy Chowdhury Prabudhya1, Zherui Han, Yixuan Sun, Dudong Feng, Guang Lin*, and Xiulin Ruan, Fast and Accurate Machine Learning of Phonon Scattering Rates and Lattice Thermal Conductivity, Nature NPJ Computational Material 9, 95, 2023.
5. Ziqi Guo, Zherui Han, Dudong Feng, Guang Lin, Xiulin Ruan, Sampling-accelerated prediction of phonon scattering rates for converged thermal conductivity and radiative properties, Nature NPJ Computational Materials, 10, 31, 2024.