MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641107156 A) filed by Hindustan Institute Of Technology And Science on September 07, 2026, for Multiple Blood Cancer Prediction From Cell Images.
Inventors include Senthil Murugan V; and Karunambikai R.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: Blood cancers are among the most deadly diseases of the blood, generating abnormalities in tissues like bone marrow and lymphatic system where they result in the production of abnormal blood cells that proliferate uncontrollably. In order to treat the disease and save the patient's life, it is extremely important that blood cancer be detected accurately and at an early stage. Traditionally, diagnosis of the disease is done through manual microscopic examination of the blood smear images which not only require a lot of time, but are also subjective and dependent on the hematologists' expertise. Hence, this study has come up with the idea of an automatic system that determines the different blood cancers from the microscopic cell images by employing deep learning methods. The suggested system can recognize four main types of blood cancers: acute lymphoblastic leukemia (ALL), acute myelogenous leukemia (AML), chronic lymphocytic leukemia (CLL), and multiple myeloma (MM). In any of these disorders, one or more of the following cellular modifications can be noticed: nuclear shape distortion, an extremely high nucleus to cytoplasm ratio, cell size variation, different texture pattern occurrences, etc. Therefore, the suggested system will be capable of automatically recognizing cancer in blood cell images, identifying the features affected by cancer, and hence, accurately classifying the types of cancer. The whole framework is built and run using Python IDLE and incorporates TensorFlow, Keras, OpenCV, NumPy, and Matplotlib as the libraries for training, testing, and performance evaluation purposes. The experimental data demonstrated that the proposed deep learning model attained a very high accuracy level in identifying cancer, determining the affected blood cells, and correctly categorizing those cells into the four types of blood cancer. The use of this automated technique not only facilitates the diagnosis process but also reduces the chances of human errors and increases the reliability of the results as compared to the conventional methods of diagnosis
Disclaimer: Curated by HT Syndication.
