SeltaSquare

1. 
2. 

 

- Secures 'Hybrid Safety Signal Detection' technology that overcomes the limitations of frequency-based analysis

- Advances signal detection by quantifying an 'ordinal severity' scale beyond simple presence/absence of adverse events, incorporating causal-inference-based XAI

- Plans to upgrade next-generation pharmacovigilance platform 'iVigilance Square' and expand CDSS supply to healthcare institutions
 

Pharmacovigilance (PV) and life-science data company SeltaSquare (CEO Shin Min-kyung) announced on the 14th that it has been finally selected for the Bio division of the "2026 Startup Growth Technology Development Program (Diddimdol – Leap-stage Deep-Tech-linked R&D)," hosted by the Ministry of SMEs and Startups.

Building on this selection, SeltaSquare will accelerate development of an explainable-AI (XAI)-based "Real-time Signal Detection" and "drug adverse-reaction causality assessment platform." Conventional frequency-based analysis methods (such as PRR and ROR) have a structural limitation: they fail to control for confounding factors such as age, underlying conditions, and concomitant medications, resulting in a proliferation of false-positive signals.

According to data from the Korea Institute of Drug Safety and Risk Management, the number of adverse drug event reports in Korea reached 277,279 in 2025 and continues to rise each year. At the same time, global regulators such as the FDA and EMA are formally requiring the submission of causal evidence based on real-world data (RWD), increasing the need for more advanced regulatory science.

The core technology SeltaSquare is developing does not classify adverse events with a simple binary distinction. Instead, it defines an ordinal severity scale of "observation–mild–severe," quantifying even the trend of escalating risk. Notably, by combining causal-inference techniques with XAI, the technology goes beyond simple statistical association to establish the actual causal relationship between a drug and its adverse reaction. Its point of differentiation is a "Hybrid Safety Signal Detection" approach that integrates statistics, machine learning, deep learning, causal inference, and XAI into a single framework.

The company plans to link the causal-inference technology secured through this R&D program to its integrated pharmacovigilance platform, iVigilance Square, to further upgrade it. Going forward, it intends to expand the business by supplying the technology to pharmaceutical companies in the form of a "causal-inference analysis module," and to primary- and tertiary-level medical institutions in the form of a "Clinical Decision Support System (CDSS)."

Kim Dong-wook, Head of SeltaSquare's Data Center, said, "Being selected for this government R&D program is a meaningful achievement that formally recognizes the technological capabilities SeltaSquare has built through the Deep-Tech Startups 1000+ project," adding, "We will continue to advance pharmacovigilance technology that combines causal inference and XAI, and take the lead in protecting public health and safety from preventable adverse drug reactions."

Meanwhile, SeltaSquare was also selected in August for the "2026 Medical AI Data Utilization Voucher Support Program," hosted by the Ministry of Health and Welfare, and is currently validating RMP-based RWD safety analysis. Through this, the company continues to expand its technological competitiveness — from R&D of AI- and RWD-based pharmacovigilance technology through to real-world validation using actual medical data.

The Original Article: 셀타스퀘어, 중기부 디딤돌 초격차 R&D 선정 < 제약·바이오 < 산업 < 기사본문 - 팜뉴스