Project

Research Area | Unorthodox Machine Learning Solutions for Cancer, Oral, and Infectious Diseases

Pratik Shah

Our published research findings listed below share key examples for reducing dependence on specialized medical devices, biological & chemical processes & creation of new paradigms for low-cost biomarker imaging and clinical diagnoses at the point-of-care. We have also demonstrated series of generative, prediction and classification algorithmic approaches for obtaining medical diagnostic information of organs & tissues from photographs captured by mobile phones and cameras. 

  • a) In collaboration with Brigham and Womens Hospital in Boston MA, we devised and published a novel “Computational staining” system to digitally stain photographs of unstained tissue biopsies with Haematoxylin and Eosin (H&E) dyes to diagnose cancer. The paper also described an automated "Computational destaining" algorithm that can remove dyes and stains from photographs of previously stained tissues, allowing reuse of patient samples. Our method uses neural networks to help physicians provide timely information about the anatomy and structure of the organ and saving time and precious biopsy samples. (Project link… View full description

Our published research findings listed below share key examples for reducing dependence on specialized medical devices, biological & chemical processes & creation of new paradigms for low-cost biomarker imaging and clinical diagnoses at the point-of-care. We have also demonstrated series of generative, prediction and classification algorithmic approaches for obtaining medical diagnostic information of organs & tissues from photographs captured by mobile phones and cameras. 

  • a) In collaboration with Brigham and Womens Hospital in Boston MA, we devised and published a novel “Computational staining” system to digitally stain photographs of unstained tissue biopsies with Haematoxylin and Eosin (H&E) dyes to diagnose cancer. The paper also described an automated "Computational destaining" algorithm that can remove dyes and stains from photographs of previously stained tissues, allowing reuse of patient samples. Our method uses neural networks to help physicians provide timely information about the anatomy and structure of the organ and saving time and precious biopsy samples. (Project link)
  •  b)  In collaboration with Beth Israel Deaconess Medical Center in Boston MA, we investigated use of dark field imaging of capillary bed under the tongue of consenting patients in emergency rooms for diagnosing sepsis (a blood borne bacterial infection). A neural network capable of distinguishing between images from non-septic and septic patients with more than 90% accuracy was reported for the first time . This approach can rapidly stratify and offer rational use of antibiotics and reduce disease burden in hospital emergency rooms and patients. (Project link)
  • c) We successfully predicted signatures associated with fluorescent porphyrin biomarkers (linked with tumors and periodontal diseases) from standard white-light photographs of the mouth, thus reducing the need for fluorescent imaging.We are expanding the repertoire of biomarkers that can be detected in RGB color images acquired at the point-of-care and pairing them with automated machine learning exams. (Project link
  •  d) We have also communicated research studies reporting automated segmentation of oral diseases from standard photographs by neural networks and correlations with systemic health conditions such as optic nerve abnormalities in patients. These examples communicate our contributions to design novel neural networks and processes that can assist physicians and patients by next-generation computational medicine algorithms at the point-of-care which integrate seamlessly into clinical workflows in hospitals all over the world. (Project link 1, Project link 2)