Dima Shulkin: Next-Generation AI Models Improve Early Sepsis Recognition
Dima Shulkin, Co-Founder and CTO at RobotDreams GmbH, shared a post on LinkedIn about a recent article by Tai-Han Lin et al, published in Journal of Medical Internet Research, adding:
“What if a routine blood test could flag a deadly infection in minutes instead of days?
A 2025 study by Lin and colleagues, published in the JMIR Publications, trained AI to spot sepsis (a life-threatening overreaction to infection that causes about 1 in 5 deaths worldwide) using only a standard blood count.
Looking back at records from 746 intensive care patients and 746 healthy outpatients in Taiwan, the best AI scored 0.90 out of 1.0 at telling sick from healthy.”
Title: AI-Driven Innovations for Early Sepsis Detection by Combining Predictive Accuracy With Blood Count Analysis in an Emergency Setting: Retrospective Study
Authors: Tai-Han Lin, Hsing-Yi Chung, Ming-Jr Jian, Chih-Kai Chang, Hung-Hsin Lin, Chiung-Tzu Yen, Sheng-Hui Tang, Pin-Ching Pan, Cherng-Lih Perng, Feng-Yee Chang, Chien-Wen Chen, Hung-Sheng Shang
Other posts featuring Dima Shulkin on Hemostasis Today.
-
Aug 23, 2026, 17:51Gurpreet Kaurr: A Proud Milestone in Haematology and Oncology
-
Aug 23, 2026, 17:33Kathrine Kaspersen: How Voluntary Plasma Donors Helped Transform Denmark’s Plasma Supply
-
Aug 23, 2026, 16:51Sthanu Subramanian: Empagliflozin and a Rise in Hematocrit
-
Aug 23, 2026, 16:50Lale Tokgözoğlu: New PCSK9i Research Published in the European Heart Journal
-
Aug 23, 2026, 16:44Eric Gehrie: Your 2026 Guide to the Essentials of Transfusion Medicine
-
Aug 23, 2026, 16:35Thomas Dayspring: The Overlooked Pathway in Cholesterol Homeostasis
-
Aug 23, 2026, 16:28Evan Mapelu: The People Who Make Cancer Care Meaningful
-
Aug 23, 2026, 16:09Felice Gragnano: New Publication on Aspirin Discontinuation After PCI for Acute Coronary Syndromes
-
Aug 23, 2026, 16:06Mostafa Faisal: Donor Type and Long-Term Quality of Life After HCT for Severe Aplastic Anemia