The results of this study suggest that a personalized antibiotic treatment time recommendation model can help doctors make better decisions, avoid potential harm from delayed treatment or premature drug delivery, while reducing patient mortality and medical costs.
Sepsis is a disease in which an excessive inflammatory response caused by infection leads to tissue damage, organ failure and death. Sepsis causes 6% hospital hospitalizations and 35% inpatient mortality, and the annual economic burden in the United States exceeds $27 billion. Currently, sepsis treatment guidelines recommend initiating broad-spectrum antibiotic treatment and completing antibiotic administration within 1 hour in all patients with suspected sepsis or septic shock.
However, recent studies suggest that this “one size fits all” treatment guideline may have serious drug side effects, such as antibiotic resistance and diarrhea, fever, nausea and abdominal pain caused by Clostridium difficile. Therefore, there is an urgent need to develop personalized antibiotic treatment timing for sepsis patients to avoid unnecessary risks.
Recently, scientists at the Ohio State University Medical Artificial Intelligence Laboratory developed a new method. They use machine learning and causal reasoning to evaluate and recommend the best treatment time for sepsis and medical decisions to use antibiotics for sepsis patients, in order to better help doctors provide personalized treatment plans to patients.
The team is committed to building a more complete and reliable clinical decision-making system that accurately provides the best antibiotic administration time. In future actual clinical deployments, based on this model, clinical treatment guidelines and doctors ‘actual clinical experience will be effectively combined to further optimize the model recommendation results.
On April 6, a related paper titled “Estimating treatment effects for time-to-treatment antibiotic stewardship in sepsis” was published in the Nature journal Nature Machine Intelligence [1].
Liu Ruoqi, a doctoral student in the Department of Computer Science and Engineering at Ohio State University, is the first author of the paper, and Zhang Ping, a dual assistant professor in the Department of Computer Science and Engineering and the Department of Biomedical Informatics at Ohio State University, is the corresponding author of the paper.
The team proposed a new method called “T4” to evaluate the duration of sepsis treatment and the effectiveness of antibiotic treatment. T4 determines the personalized optimal antibiotic treatment time by analyzing the patient’s historical observation data, controlling dynamic confounding variables, and evaluating individual potential outcomes and causal effects corresponding to different treatment times in the future.
Relevant experiments used real-world patient data in the United States and Europe to evaluate the accuracy of individual causal effects and the effectiveness of the model’s recommended antibiotic treatment timing.
If you want to learn more, you can click on the link below the video.
Thank you for watching this video. If you like it, please subscribe and like it. thank
Original text:https://www.mittrchina.com/news/detail/12056
Video: