MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641107793 A) filed by Pragati Engineering College on September 08, 2026, for Image-Assisted Differential Equation Modeling Of Leukemia Dynamics Using Transfer Learning- Derived Morphological Features.
Inventors include Dr. Avinash Gudimetla; Mrs. Aravinda Karri; Mr. N Raghuveer; and Mr. V V N Sarath.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: This paper develops a mathematical framework that couples a compartmental system of nonlinear ordinary differential equations (ODEs), describing the competitive dynamics of healthy hematopoietic cells, leukemic blasts, an effector T-cell population, and a regulatory cytokine, with a scalar morphological aggressiveness score obtained from transfer-learning based feature extraction applied to peripheral blood smear images. The governing equations generalize established leukemia population- dynamics models to admit an image-derived, patient-specific modulation of the leukemic effective growth rate. We state the underlying assumptions explicitly, derive the governing equations, discuss their physical interpretation, and numerically integrate the system in MATLAB across a range of morphology scores. The simulations reproduce qualitatively realistic phenomena reported in the mathematical oncology literature, including slow indolent growth, immune-mediated relapse–remission oscillations, and a sharply accelerated approach to a clinically defined blast-crisis threshold as the image-derived aggressiveness score increases. We discuss applications, limitations, and recent developments at the intersection of deep learning and differential-equation modeling of hematological malignancies. All equations, assumptions, and quantitative claims are supported by citation to peer-reviewed sources; statements that could not be corroborated in the available literature are explicitly flagged.
Disclaimer: Curated by HT Syndication.
