MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641073936 A) filed by N J R Muniraj; Saravana Sundaram; M. Deeparani; T. Sivakumar; Shobha Christila; R. Vishw Allth; Murugeswari T; K. Kiruthika; R. Gayathri Priyadarshini; Afreen A; Prashanth S; and M. Sreelakshmi on June 15, 2026, for Custom Low Power Application Specific Integrated Circuit For Ai Driven Volatile Organic Compound Classification In Invasive Cancer Screening Using Cpdk 90 Nm Technology.
Inventors include N J R Muniraj; S. Saravana Sundaram; M. Deeparani; T. Sivakumar; S. Shobha Christila; R. Vishw Allth; Murugeswari; K. Kiruthika; R. Gayathri Priyadarshini; S. Afreena; Prashanth S; and M. Sreelakshmi.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: This invention presents a novel ultra-low-power custom Application-Specific Integrated Circuit (Breathomics-ASIC) for real-time processing of Volatile Organic Cornpuurul (VOC) signatures from an 8-channel Metal-Oxide Semiconductor (MOS) nanosensor array, targeted at_ non-invasive early detection of lung and oral cancers. The ASIC core integrates multichannel ADC data acquisition with double-buffering, DAC calibration control, fixed-point tisk score computation engine, and per-sensor power gating in a highly optimized digital design. The complete RTL-to-synthesis flow was executed using Cadence EDA tools, with functional simulation in Sim Vision and logic synthesis in Genus 20.1 targeting the GPDK 90nm standard cell library. The design operates at 200 MHz (5 ns clock period) and achieves excellent timing closure with a positive slack of +4,709 ps. Post-synthesis results confirm a compact implementation with only 113 standard cells, total power consumption of 172.708 Jl W, and full functional verification of data acquisition, calibration, risk scoring, and power management. The Breathomics-ASIC serves as the core hardware accelerator within a dual-pathway AI platform employing Random Forest ensemble learning and 1 D-CNN with late fusion, enabling sub-200 JlW edge preprocessing for portable, battery-operated point-of-care breathomics devices. The invention is ready for physical design, signoff, and tape-out in GPDK 90nm technology, making it highly suitable for scalable, affordable cancer screening solutions in resource-constrained settings.
Disclaimer: Curated by HT Syndication.
