Estimating Surgical Blood Loss Volume Using Continuously Monitored Vital Signs

Automated deep learning model for estimating intraoperative blood loss using gauze images

The intraoperative estimated blood loss (EBL), an essential parameter for perioperative management, has been evaluated by manually weighing blood in gauze and suction bottles, a process both time-consuming and labor-intensive. As the novel EBL prediction platform, we developed an automated deep learning EBL prediction model, utilizing the patch-wise crumpled state (P-W CS) of gauze images with texture analysis. The proposed algorithm was developed using animal data obtained from a porcine experiment and validated on human intraoperative data prospectively collected from 102 laparoscopic gastric cancer surgeries. The EBL prediction model involves gauze area detection and subsequent EBL regression based on the detected areas, with each stage optimized through comparative model performance evaluations. The selected gauze detection model demonstrated a sensitivity of 96.5% and a specificity of 98.0%. Based on this detection model, the performance of EBL regression stage models was compared. Comparative evaluations revealed that our P-W CS-based model outperforms others, including one reliant on convolutional neural networks and another analyzing the gauze's overall crumpled state. The P-W CS-based model achieved a mean absolute error (MAE) of 0.25 g and a mean absolute percentage error (MAPE) of 7.26% in EBL regression. Additionally, per-patient assessment yielded an MAE of 0.58 g, indicating errors < 1 g/patient. In conclusion, our algorithm provides an objective standard and streamlined approach for EBL estimation during surgery without the need for perioperative approximation and additional tasks by humans. The robust performance of the model across varied surgical conditions emphasizes its clinical potential for real-world application.

Development of an Anticipatory Triage-Ranking Algorithm Using Dynamic Simulation of the Expected Time Course of Patients With Trauma: Modeling and Simulation Study

BACKGROUND

In cases of terrorism, disasters, or mass casualty incidents, far-reaching life-and-death decisions about prioritizing patients are currently made using triage algorithms that focus solely on the patient's current health status rather than their prognosis, thus leaving a fatal gap of patients who are under- or overtriaged.

OBJECTIVE

The aim of this proof-of-concept study is to demonstrate a novel approach for triage that no longer classifies patients into triage categories but ranks their urgency according to the anticipated survival time without intervention. Using this approach, we aim to improve the prioritization of casualties by respecting individual injury patterns and vital signs, survival likelihoods, and the availability of rescue resources.

METHODS

We designed a mathematical model that allows dynamic simulation of the time course of a patient's vital parameters, depending on individual baseline vital signs and injury severity. The 2 variables were integrated using the well-established Revised Trauma Score (RTS) and the New Injury Severity Score (NISS). An artificial patient database of unique patients with trauma (N=82,277) was then generated and used for analysis of the time course modeling and triage classification. Comparative performance analysis of different triage algorithms was performed. In addition, we applied a sophisticated, state-of-the-art clustering method using the Gower distance to visualize patient cohorts at risk for mistriage.

RESULTS

The proposed triage algorithm realistically modeled the time course of a patient's life, depending on injury severity and current vital parameters. Different casualties were ranked by their anticipated time course, reflecting their priority for treatment. Regarding the identification of patients at risk for mistriage, the model outperformed the Simple Triage And Rapid Treatment's triage algorithm but also exclusive stratification by the RTS or the NISS. Multidimensional analysis separated patients with similar patterns of injuries and vital parameters into clusters with different triage classifications. In this large-scale analysis, our algorithm confirmed the previously mentioned conclusions during simulation and descriptive analysis and underlined the significance of this novel approach to triage.

CONCLUSIONS

The findings of this study suggest the feasibility and relevance of our model, which is unique in terms of its ranking system, prognosis outline, and time course anticipation. The proposed triage-ranking algorithm could offer an innovative triage method with a wide range of applications in prehospital, disaster, and emergency medicine, as well as simulation and research.

Intelligent Clinical Decision Support

Early recognition of pathologic cardiorespiratory stress and forecasting cardiorespiratory decompensation in the critically ill is difficult even in highly monitored patients in the Intensive Care Unit (ICU). Instability can be intuitively defined as the overt manifestation of the failure of the host to adequately respond to cardiorespiratory stress. The enormous volume of patient data available in ICU environments, both of high-frequency numeric and waveform data accessible from bedside monitors, plus Electronic Health Record (EHR) data, presents a platform ripe for Artificial Intelligence (AI) approaches for the detection and forecasting of instability, and data-driven intelligent clinical decision support (CDS). Building unbiased, reliable, and usable AI-based systems across health care sites is rapidly becoming a high priority, specifically as these systems relate to diagnostics, forecasting, and bedside clinical decision support. The ICU environment is particularly well-positioned to demonstrate the value of AI in saving lives. The goal is to create AI models embedded in a real-time CDS for forecasting and mitigation of critical instability in ICU patients of sufficient readiness to be deployed at the bedside. Such a system must leverage multi-source patient data, machine learning, systems engineering, and human action expertise, the latter being key to successful CDS implementation in the clinical workflow and evaluation of bias. We present one approach to create an operationally relevant AI-based forecasting CDS system.

Automated Assessment of Cardiovascular Sufficiency Using Non-Invasive Physiological Data

For fluid resuscitation of critically ill individuals to be effective, it must be well calibrated in terms of timing and dosages of treatments. In current practice, the cardiovascular sufficiency of patients during fluid resuscitation is determined using primarily invasively measured vital signs, including Arterial Pressure and Mixed Venous Oxygen Saturation (SvO2), which may not be available in outside-of-hospital settings, particularly in the field when treating subjects injured in traffic accidents or wounded in combat where only non-invasive monitoring is available to drive care. In this paper, we propose (1) a Machine Learning (ML) approach to estimate the sufficiency utilizing features extracted from non-invasive vital signs and (2) a novel framework to address the detrimental impact of inter-patient diversity on the ability of ML models to generalize well to unseen subjects. Through comprehensive evaluation on the physiological data collected in laboratory animal experiments, we demonstrate that the proposed approaches can achieve competitive performance on new patients using only non-invasive measurements. These characteristics enable effective monitoring of fluid resuscitation in real-world acute settings with limited monitoring resources and can help facilitate broader adoption of ML in this important subfield of healthcare.