The Mehanian lab develops machine learning algorithms to solve healthcare problems relating to the detection and diagnosis of disease or the prognosis of health risks. Diagnostic functions are traditionally carried out by highly trained medical professionals, but there are drawbacks to relying exclusively on human judgment, which is characterized by high intra- and inter-person variability and varying levels of skill. Machine learning algorithms are trained with large, carefully curated, expertly annotated, high-quality datasets to perform these tasks at performance levels comparable to highly-trained medical professionals, and with greater consistency. Through the use of AI-enabled systems, expert-level decision support becomes available in any environment, but is particularly valuable in austere or low-resource settings, which may suffer from a paucity of trained medical personnel. Example applications include microscopy, ultrasound, X-ray, CT, and MRI.

We are looking for people who are excited about working on artificial intelligence in bio-medicine. If you are trained in the life sciences and have a desire to acquire machine learning skills and apply them to bio-medical problems, our lab may be the place for you. If you are a computer scientist who is keenly interested in biomedical applications of machine learning, our lab may be the place for you.

We welcome people of all backgrounds!

*Images reproduced with permission of Global Health Labs, LLC and Oregon Health & Science University.
Upper-Left: Mehanian, et. al, “Artificial Intelligence-based Automated Interpretation of Lung Ultrasound Images of Pneumothorax, Pleural Effusion and Pneumonia: Results from Human Studies,” 2019 Military Health System Research Symposium.
Upper-Right: Horning et. al, “Limitations of Hemozoin as a Diagnostic Biomarker for Malaria,” 2014 IEEE Global Humanitarian Technology Conference.
Bottom Two: Hu et al. “An Observational Study of Deep Learning and Automated Evaluation of Cervical Images for Cancer Screening”, Journal of National Cancer Institute, vol. 111 (9), 2019.