Scientists at the Siberian State Medical University (SibSMU) have developed a highly accurate system for rapidly assessing blood loss in ambulances and intensive care units. The findings of this groundbreaking research were unveiled at the International Scientific and Technical Conference for Young Scientists and Students, held in Minsk to commemorate the 105th anniversary of the Belarusian National Technical University.
Prompt detection of hemorrhage is crucial for saving lives, especially in disaster medicine scenarios where rapid assessment of casualties is essential, such as in various accidents, road traffic incidents, and natural disasters.
According to SibSMU researchers, while modern medical monitors provide real-time physiological data for patient assessment, the accuracy of this assessment heavily relies on the experience and qualifications of medical staff.
To mitigate human error and automate the blood loss assessment process, the university`s specialists developed an innovative algorithm. This algorithm aims to create a cyber-physical system model for monitoring patient conditions in emergency vehicles and intensive care units.
The core of this innovation involves modeling hemorrhagic shock using a modified version of the Pulse Physiology Platform, a human physiology simulator.
University researchers state that a neural network is employed to process data, analyzing key indicators such as arterial blood pressure, heart rate, oxygen saturation, and other vital parameters.
“At the current stage, the algorithm has been tested, and the results have satisfied specialists from the resuscitation and intensive care unit,” said Gleb Vishtalyuk, a SibSMU student.
He added that a review of scientific literature indicates no existing analogues of such a system in global medical practice.
The proposed algorithm demonstrated high efficacy, achieving 99 percent accuracy in determining blood loss based on key physiological indicators. The system was trained using a robust database containing over 21,840 physiological records.
Currently, scientists are focused on validating the developed model using real data from intensive care units. To facilitate this, SibSMU clinics have integrated a resuscitation and anesthesiology information system, which allows for the continuous collection of patient data within these units.
This significant innovation was developed at the “Digital Medicine and Cyberphysics” scientific and technological center, established under the “Priority-2030” national development program.
Rupert Blackwood 
Clement Ashworth