DifferAI Integrates Edge AI Semiconductor into SynergyAI Medical AI Solution

Artificial intelligence (AI) fabless semiconductor design company DifferAI announced on the 5th that it has successfully integrated its domestically produced edge AI semiconductor 'Tachy-BS402', which recently achieved mass production, into the arrhythmia prediction diagnostic solution 'Mac’AI' developed by Synergy AI. Synergy AI is a medical AI software specialist company founded by Professor Shin Tae-young of the Department of Urology at Ewha Womans University Mokdong Hospital.

DifferAI's edge AI semiconductor 'Tachy-BS402', embedded in Synergy AI's Mac’AI, is designed to run deep learning-trained AI models on an ultra-compact USB-type semiconductor. The embedded model demonstrated excellent performance, showing a prediction accuracy of 92.7% for various arrhythmias. The successful integration of the edge AI semiconductor into Mac’AI represents a breakthrough in reducing power consumption compared to existing GPU-based cloud systems, achieving top-level data security and maximizing cost savings, making it an exemplary case of AI semiconductor commercialization.

This achievement was decisively supported by the AI Semiconductor Application Demonstration Support Project under the National IT Industry Promotion Agency (NIPA), affiliated with the Ministry of Science and ICT.

A Synergy AI representative stated, “The successful integration of the edge AI-NPU into the Mac’AI solution, which enables innovative predictive treatment for arrhythmia diagnosis, is expected to play a pivotal role as a catalyst for the growth of the domestic AI semiconductor industry. We have completed the domestic MFDS confirmatory clinical trials, and once the prescribed procedures such as innovative medical device registration are finalized, this could provide an opportunity for South Korea to enter commercialization faster than the United States.”

He added, “We have gained confidence that by moving away from the existing cloud-based operation method and utilizing DifferAI’s edge AI semiconductor, we can expand various business models.”

A DifferAI representative emphasized, “With substantial support from the AI Semiconductor Application Demonstration Support Project under NIPA, affiliated with the Ministry of Science and ICT, we succeeded in integrating AI semiconductors into medical AI solutions. Although most AI solutions adopt cloud-based services, there are limitations in the application process such as the high cost of local server construction, power grid reconfiguration, and cloud network usage.”

He continued, “We expect the successful case of embedding a domestically produced AI semiconductor-based model to have a significant impact on the entire AI industry. In addition, we are developing various solutions to apply AI semiconductors to AI services in diverse fields.”

DifferAI and Synergy AI signed a business agreement in March, and Mac’AI is reportedly set to receive the Innovation Award at the International Consumer Electronics Show (CES), the world’s largest electronics exhibition, in recognition of its service excellence.

Meanwhile, the medical AI market is expected to grow rapidly as applying AI technology to the pharmaceutical industry not only accelerates the drug discovery process but also enhances the efficacy of commercialized treatments. According to a report by the Korea Health Industry Development Institute, the total expenditure on AI technology in the global pharmaceutical and bio industries is expected to exceed $3 billion (approximately 3.9 trillion KRW) by 2025.

Based on various advantages, investment and collaboration cases between global pharmaceutical companies and AI firms are steadily increasing. Boehringer Ingelheim signed a platform development agreement with global IT company IBM to discover new antibody candidates and last month signed a technology transfer contract worth $500 million (approximately 650 billion KRW) with Canadian AI bio company PhenomicAI. Additionally, companies such as Novo Nordisk and Sanofi are also engaging in AI-driven new drug development.

This content was produced with the assistance of AI translation services.