MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641108928 A) filed by Ms. Nidhi Saraswat; Ms. Bairavi S M; Ms. Rekha S; Dr Deepak Tandon; Dr. Reshu Gupta; Dr. Reshma J; Ms. Pavithra J; Dr Sandhya Divakar Moolbharti; Prof. Dr. S. P. Singh; and Dr. Sharmila Arunkumar on September 10, 2026, for Artificial Intelligence Assisted Predictive Healthcare Diagnostic Device Platform.
Inventors include Ms. Nidhi Saraswat; Ms. Bairavi S M; Ms. Rekha S; Dr Deepak Tandon; Dr. Reshu Gupta; Dr. Reshma J; Ms. Pavithra J; Dr Sandhya Divakar Moolbharti; Prof. Dr. S. P. Singh; and Dr. Sharmila Arunkumar.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: ABSTRACT An artificial-intelligence-assisted predictive healthcare diagnostic device platform is disclosed. The platform receives synchronized physiological, biochemical, imaging, behavioural, and clinical-context data from integrated or connected sources. A preprocessing engine determines provenance, calibration, missingness, and quality for each stream. A patient-state encoder generates longitudinal features including deviation from an individual baseline. Multiple predictive models process structured, temporal, image, or audio data, and a fusion controller combines only available validated modalities according to quality. An uncertainty estimator provides calibrated confidence and detects distribution shift. A deterministic clinical-safety engine modifies, suppresses, or escalates the predictive output using physiologic rules, estimated severity, time-to-event, uncertainty, and patient-specific context. An authorized interface presents a ranked risk or candidate condition together with contributing factors, confidence limitations, and a recommended confirmatory action. Edge inference permits operation during network loss, while consent control, encryption, signed model packages, audit logging, clinician feedback, and drift monitoring support secure and traceable deployment. The platform thereby provides early, explainable, confidence-aware clinical decision support without removing professional oversight. Statement Regarding Scope The foregoing embodiments illustrate the inventive concept and are not restrictive. Components may be distributed between a patient device, trusted gateway, and clinical server, provided that the claimed quality-aware multimodal prediction, uncertainty estimation, safety control, explainability, and traceability are retained. The scope of protection is determined by the appended claims.
Disclaimer: Curated by HT Syndication.
