Continuous Temperature Measurement by Thermal Imaging Camera: Concordance, Patterns, and Intelligent Prediction of Events in Critical Patients

This study delves into the nuanced domain of body temperature monitoring in the context of hospitalized patients, with a particular emphasis on critical care settings. The primary physiological variable under scrutiny is body temperature, a parameter that has been extensively examined in the clinical realm. It is underscored that the normal range for body temperature hovers around 36.7°C, but with a significant degree of variability spanning from 35.3 to 37.7°C, both among different individuals and within the same person throughout the day.

Fever, a crucial clinical manifestation, is operationally defined by a consensus reached by the American College of Critical Care Medicine and the Infectious Diseases Society of America, placing it at a central temperature of 38.3°C. This central temperature refers to the temperature of internal organs, and the thermal elevation associated with fever is not exclusive to infectious processes but extends to various other conditions, including autoimmune diseases, oncological conditions, bleeding, inflammatory reactions, surgical procedures, and drug-induced scenarios.

The multifaceted landscape of temperature measurement methods in Intensive Care Units (ICUs) is acknowledged, with peripheral temperature measurements using contact thermometers in the axilla being the most common approach. For patients with sustained fever or those undergoing therapeutic hypothermia, continuous monitoring of central temperature through various methods such as rectal, tympanic, vesical, or esophageal measurements is considered. However, the more invasive nature of central temperature monitoring, coupled with technical challenges, higher economic costs, and potential complications, limits its widespread utilization.

The critical nature of thermal monitoring in the care of patients is highlighted, especially considering that fever often prompts antibiotic treatment, even in the absence of confirmed infection, due to the observed improvement in survival rates among septic patients with early intervention. The complex landscape surrounding temperature measurement, characterized by individual variability, circadian changes, diverse measurement methods, medication influence, and methodological deficiencies, prompts a reconsideration of the feasibility of defining fever based on a single temperature value. To address this complexity, the study introduces the concept of the thermal curve, representing the temporal evolution of temperature throughout the day, which may exhibit sustained elevation or peaks and may or may not respond to interventions like antipyretics, antibiotics, or other temperature control methods.

An additional challenge in fever monitoring, particularly with sporadic measurements rather than continuous monitoring, is the potential oversight of febrile peaks or the failure to capture the maximum or minimum temperature values experienced by the patient. This limitation underscores the need for innovative approaches to temperature monitoring that can overcome these challenges.