Pratik Shah presented two research papers at the 2018 17th IEEE International Conference on Machine Learning and Application. These papers, reporting novel AI and machine learning methods to help physicians in pathology clinics and hospital emergency rooms, explore how these neural networks work synergistically with physicians at the point-of-care by either automating time-consuming and expensive biological and clinical steps (such as staining of tissue biopsies) or by providing rapid diagnostic aid for complex diseases (such as sepsis).
- In the first publication, presented as a short paper, Shah describes a “computational staining” approach to digitally stain photographs of unstained tissue biopsies with Haematoxylin and Eosin (H&E) dyes to diagnose cancer. Their method uses neural networks to rapidly stain photographs of non-stained tissues, providing physicians timely information about the anatomy and structure of the tissue and saving both time and biopsy samples (which are limited in number). The paper also describes an automated "computational destaining" algorithm that can remove dyes and stains from photographs of previously stained tissues, allowing reuse of patient samples.
- In the second paper, presented as an invited talk, Shah investigates 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 is reported for the first time. This research can rapidly stratify and offer rational use of antibiotics and reduce disease burden in hospital emergency rooms and patients.
Both these publications describe neural networks that can assist physicians and patients by novel computational processes at the point-of-care and integrate seamlessly into clinical workflows in hospitals all over the world.