Emam, Omar S.

Introduction A steadily rising opioid pandemic has left the US suffering significant social, economic, and health crises. Machine learning (ML) domains have been utilized to predict prolonged postoperative opioid (PPO) use. This systematic review aims to compile all up-to-date studies addressing such algorithms’ use in clinical practice. Methods We searched PubMed/MEDLINE, EMBASE, CINAHL, and Web of Science using the keywords “machine learning,” “opioid,” and “prediction.” The results were limited to human studies with full-text availability in English. We included all peer-reviewed journal articles that addressed an ML model to predict PPO use by adult patients. Results Fifteen studies were included with a sample size ranging from 381 to 112898, primarily orthopedic-surgery-related. Most authors define a prolonged misuse of opioids if it extends beyond 90 days postoperatively. Input variables ranged from 9 to 23 and were primarily preoperative. Most studies developed and tested at least two algorithms and then enhanced the best-performing model for use retrospectively on electronic medical records. The best-performing models were decision-tree-based boosting algorithms in 5 studies with AUC ranging from .81 to .66 and Brier scores ranging from .073 to .13, followed second by logistic regression classifiers in 5 studies. The topmost contributing variable was preoperative opioid use, followed by depression and antidepressant use, age, and use of instrumentation. Conclusions ML algorithms have demonstrated promising potential as a decision-supportive tool in predicting prolonged opioid use in post-surgical patients. Further validation studies would allow for their confident incorporation into daily clinical practice.

Pain assessment is a complex task largely dependent on the patient’s self-report. Artificial intelligence (AI) has emerged as a promising tool for automating and objectifying pain assessment through the identification of pain-related facial expressions. However, the capabilities and potential of AI in clinical settings are still largely unknown to many medical professionals. In this literature review, we present a conceptual understanding of the application of AI to detect pain through facial expressions. We provide an overview of the current state of the art as well as the technical foundations of AI/ML techniques used in pain detection. We highlight the ethical challenges and the limitations associated with the use of AI in pain detection, such as the scarcity of databases, confounding factors, and medical conditions that affect the shape and mobility of the face. The review also highlights the potential impact of AI on pain assessment in clinical practice and lays the groundwork for further study in this area.

In 2020, Mayo Clinic established an Advanced Care at Home (ACH) program. ACH is a virtual hybrid hospital-at-home (HaH) program that combines telemedicine with in-home care services by utilizing a state that is software-driven, vendor-mediate medical supply chain. The program initially focused on acute medical diagnosis but has expanded to oversee surgical and postsurgical patients with continued inpatient needs. Here, we report the first case of a small bowel obstruction (SBO) managed under a HaH program. A 52-year-old lady presented to the emergency department with symptoms suggestive of mechanical SBO. The diagnosis was confirmed with an abdominopelvic computed tomography (CT) scan, and the patient was admitted to the hospital. Based on the patient’s presentation and laboratory results, the care team proceeded with conservative treatment including nasogastric tube (NG) placement and suctioning, intravenous (IV) fluid replacement, and daily laboratory studies. She spent the first hospital day in the physical hospital ward so that the surgical team could ensure stability clinically and no urgent need for surgical intervention. On hospital day two, she was transferred home with ACH where the NG suctioning and IV replacement therapy could continue, while the medical team conducted daily virtual visits to ensure continued improvement. Additionally, a paramedic and a nurse performed an in-person, head-to-toe assessment and administered medications to the patient twice daily. She spent 5 days in ACH getting acute care and then was discharged into a postacute phase equivalent to outpatient monitoring called the restorative phase. She was monitored remotely for the duration of the restorative phase for 10 more days, and then she recovered fully. This case highlights that high-acuity patients with SBO can receive invasive treatments like NG tube suction as well as be appropriately monitored for clinical decompensation by a virtual hybrid home hospital program which combines virtual care providers with an in-home vendor-mediated supply chain.

Upper and lower limb amputations are frequently associated with phantom limb pain (PLP). Recently, virtual reality (VR) and augmented reality (AR) have been reported as a potential therapy of PLP. We have conducted a systematic review of literature to evaluate the efficacy of VR and AR in managing PLP. Four databases were searched: PubMed, EMBASE, Cumulative Index to Nursing and Allied Health Literature, and Web of Science. We utilized the Preferred Reporting Items for Systematic Reviews and Meta-Analysis for our organization. The initial search resulted in 164 results. After title, abstract, and full-text screening, 9 studies were included. One study was of good quality and 8 studies were of fair to poor quality. Seven studies utilized VR and 2 studies utilized AR. The number of treatment sessions ranged from 1 to 28 and the duration ranged from 10 minutes to 2 hours. Several pain scales were used to evaluate PLP pre- and postintervention including Numeric Rating Scale, Pain Rating Index, McGill Pain Questionnaire, and Visual Analog Scale. All the studies reported improvement of PLP on one or more of pain scales after one or more sessions of VR or AR. Despite the promising results reported by literature, we cannot recommend using VR or AR for PLP. Most of the studies are of poor design and have limited sample size with high bias levels. Therefore, no substantial evidence can be derived from them. However, we do believe further research with high-quality randomized controlled trials should take place to increase the knowledge of the potential advantages.