Algor-ethics: charting the ethical path for AI in critical care.

The integration of Clinical Decision Support Systems (CDSS) based on artificial intelligence (AI) in healthcare is groundbreaking evolution with enormous potential, but its development and ethical implementation, presents unique challenges, particularly in critical care, where physicians often deal with life-threating conditions requiring rapid actions and patients unable to participate in the decisional process. Moreover, development of AI-based CDSS is complex and should address different sources of bias, including data acquisition, health disparities, domain shifts during clinical use, and cognitive biases in decision-making. In this scenario algor-ethics is mandatory and emphasizes the integration of 'Human-in-the-Loop' and 'Algorithmic Stewardship' principles, and the benefits of advanced data engineering. The establishment of Clinical AI Departments (CAID) is necessary to lead AI innovation in healthcare, ensuring ethical integrity and human-centered development in this rapidly evolving field.

Artificial Intelligence in Operating Room Management.

This systematic review examines the recent use of artificial intelligence, particularly machine learning, in the management of operating rooms. A total of 22 selected studies from February 2019 to September 2023 are analyzed. The review emphasizes the significant impact of AI on predicting surgical case durations, optimizing post-anesthesia care unit resource allocation, and detecting surgical case cancellations. Machine learning algorithms such as XGBoost, random forest, and neural networks have demonstrated their effectiveness in improving prediction accuracy and resource utilization. However, challenges such as data access and privacy concerns are acknowledged. The review highlights the evolving nature of artificial intelligence in perioperative medicine research and the need for continued innovation to harness artificial intelligence's transformative potential for healthcare administrators, practitioners, and patients. Ultimately, artificial intelligence integration in operative room management promises to enhance healthcare efficiency and patient outcomes.

Strategies to build and maintain competence in pain management: Insights from a SIAARTI survey on educational needs among Italian anesthesiologists.

PURPOSE: Fulfilling educational needs in pain management should be a lifelong process, even involving physicians board certified in pain medicine such as the anesthesiologists/pain therapists. The aim of the study was to investigate Italian anesthesiologists' self-perceived competency, confidence, and interest to attend educational programs in relation to their seniority in pain management. METHODS: SIAARTI members were sent an online questionnaire addressing the following items: education, skills (both soft and hard skills), technical expertise and engaged to participate between December 2020 and January 2021. Participants rated their competence based on the following range (no knowledge, knowledge, competence) while their agreement to attend educational courses was assessed using a 5-point Likert-type scale. RESULTS: Less than one in four participants declare to be dedicated to pain medicine activity with greater proportion among older (over 61 years) compared to younger ones (31-40 years). Regarding cancer and chronic noncancer pain a positive gradient of self-perceived competence has been observed in relation to seniority. In contrast, no gradient of self-perceived competence was reported about musculoskeletal and low back pain. Participants self-perceived competent in both opioid use and prevention of opioid-related adverse event while feeling less competent when managing drugs with abuse potential. The lowest competence has been observed in pediatric pain along with the lowest interest to attend educational courses. Participants were much and very much interested to education regarding cancer, noncancer, musculoskeletal, and low back pain, invasive analgesic procedures but less regarding items for which they declared less competence, such as use of pain scales, pain management in children, and use of drugs with abuse potential. CONCLUSION: This work provides first evidence of a summative assessment of competency and related educational needs' profile of anesthesiologists/pain therapists thus paving the way for developing a nationwide educational program to improve chronic pain care in Italy.

Level of Diffusion and Training of Lung Ultrasound during the COVID-19 Pandemic - A National Online Italian Survey (ITALUS) from the Lung Ultrasound Working Group of the Italian Society of Anesthesia, Analgesia, Resuscitation, and Intensive Care (SIAARTI).

PURPOSE: The goal of this survey was to describe the use and diffusion of lung ultrasound (LUS), the level of training received before and during the COVID-19 pandemic, and the clinical impact LUS has had on COVID-19 cases in intensive care units (ICU) from February 2020 to May 2020. MATERIALS AND METHODS: The Italian Lung Ultrasound Survey (ITALUS) was a nationwide online survey proposed to Italian anesthesiologists and intensive care physicians carried out after the first wave of the COVID-19 pandemic. It consisted of 27 questions, both quantitative and qualitative. RESULTS: 807 responded to the survey. The median previous LUS experience was 3 years (IQR 1.0-6.0). 473 (60.9 %) reported having attended at least one training course on LUS before the COVID-19 pandemic. 519 (73.9 %) reported knowing how to use the LUS score. 404 (52 %) reported being able to use LUS without any supervision. 479 (68.2 %) said that LUS influenced their clinical decision-making, mostly with respect to patient monitoring. During the pandemic, the median of patients daily evaluated with LUS increased 3-fold (p < 0.001), daily use of general LUS increased from 10.4 % to 28.9 % (p < 0.001), and the daily use of LUS score in particular increased from 1.6 % to 9.0 % (p < 0.001). CONCLUSION: This survey showed that LUS was already extensively used during the first wave of the COVID-19 pandemic by anesthesiologists and intensive care physicians in Italy, and then its adoption increased further. Residency programs are already progressively implementing LUS teaching. However, 76.7 % of the sample did not undertake any LUS certification.

Lung Ultrasound in COVID-19 Pneumonia: Correlations with Chest CT on Hospital admission.

BACKGROUND: Lung ultrasound (LUS) is an accurate, safe, and cheap tool assisting in the diagnosis of several acute respiratory diseases. The diagnostic value of LUS in the workup of coronavirus disease-19 (COVID-19) in the hospital setting is still uncertain. OBJECTIVES: The aim of this observational study was to explore correlations of the LUS appearance of COVID-19-related pneumonia with CT findings. METHODS: Twenty-six patients (14 males, age 64 ± 16 years) urgently hospitalized for COVID-19 pneumonia, who underwent chest CT and bedside LUS on the day of admission, were enrolled in this observational study. CT images were reviewed by expert chest radiologists, who calculated a visual CT score based on extension and distribution of ground-glass opacities and consolidations. LUS was performed by clinicians with certified competency in thoracic ultrasonography, blind to CT findings, following a systematic approach recommended by ultrasound guidelines. LUS score was calculated according to presence, distribution, and severity of abnormalities. RESULTS: All participants had CT findings suggestive of bilateral COVID-19 pneumonia, with an average visual scoring of 43 ± 24%. LUS identified 4 different possible -abnormalities, with bilateral distribution (average LUS score 15 ± 5): focal areas of nonconfluent B lines, diffuse confluent B lines, small subpleural microconsolidations with pleural line irregularities, and large parenchymal consolidations with air bronchograms. LUS score was significantly correlated with CT visual scoring (r = 0.65, p < 0.001) and oxygen saturation in room air (r = -0.66, p < 0.001). CONCLUSION: When integrated with clinical data, LUS could represent a valid diagnostic aid in patients with suspect COVID-19 pneumonia, which reflects CT findings.

Cardiac arrest and cardiopulmonary resuscitation in the next decade: Predicting and shaping the impact of technological innovations.

INTRODUCTION: Cardiac arrest (CA) is the third leading cause of death, with persistently low survival rates despite medical advancements. This article evaluates the potential of emerging technologies to enhance CA management over the next decade, using predictions from the AI tools ChatGPT-4 and Gemini Advanced. METHODS: We conducted an exploratory literature review to envision the future of cardiopulmonary arrest (CA) management. Utilizing ChatGPT-4 and Gemini Advanced, we predicted implementation timelines for innovations in early recognition, CPR, defibrillation, and post-resuscitation care. We also consulted the AI to assess the consistency and reproducibility of the predictions. RESULTS: We extrapolate that healthcare may embrace new technologies, such as comprehensive monitoring of vital signs to activate the emergency system (wireless detectors, smart speakers, and wearable devices), use new innovative early CPR and early AED devices (robot CPR, wearable AEDs, and immersive reality), and post-resuscitation care monitoring (brain-computer interface). These technologies could enhance timely life-saving interventions for cardiac arrest. However, there are many ethical and practical challenges, particularly in maintaining patient privacy and equity. The two AI tools made different predictions, with a horizon for implementation ranging between three and eight years. CONCLUSION: Integrating advanced monitoring technologies and AI-driven tools offers hope in improving CA management. A balanced approach involving rigorous scientific validation and ethical oversight is necessary. Collaboration among technologists, medical professionals, ethicists, and policymakers is crucial to use these innovations ethically to reduce CA incidence and enhance outcomes. Further research is needed to enhance the reliability of AI predictive capabilities.

Maladaptive personality traits are associated with burnout risk in Italian anesthesiologists and intensivists: a secondary analysis from a cross-sectional study.

BACKGROUND: Burnout is a maladaptive response to chronic stress, particularly prevalent among clinicians. Anesthesiologists are at risk of burnout, but the role of maladaptive traits in their vulnerability to burnout remains understudied. METHODS: A secondary analysis was performed on data from the Italian Association of Hospital Anesthesiologists, Pain Medicine Specialists, Critical Care, and Emergency (AAROI-EMAC) physicians. The survey included demographic data, burnout assessment using the Maslach Burnout Inventory (MBI) and subscales (emotional exhaustion, MBI-EE; depersonalization, MBI-DP; personal accomplishment, MBI-PA), and evaluation of personality disorders (PDs) based on DSM-IV (Diagnostic and Statistical Manual of Mental Disorders Fourth Edition) criteria using the assessment of DSM-IV PDs (ADP-IV). We investigated the aggregated scores of maladaptive personality traits as predictor variables of burnout. Subsequently, the components of personality traits were individually assessed. RESULTS: Out of 310 respondents, 300 (96.77%) provided complete information. The maladaptive personality traits global score was associated with the MBI-EE and MBI-DP components. There was a significant negative correlation with the MBI-PA component. Significant positive correlations were found between the MBI-EE subscale and the paranoid (r = 0.42), borderline (r = 0.39), and dependent (r = 0.39) maladaptive personality traits. MBI-DP was significantly associated with the passive-aggressive (r = 0.35), borderline (r = 0.33), and avoidant (r = 0.32) traits. Moreover, MBI-PA was negatively associated with dependent (r = - 0.26) and avoidant (r = - 0.25) maladaptive personality features. CONCLUSIONS: There is a significant association between different maladaptive personality traits and the risk of experiencing burnout among anesthesiologists. This underscores the importance of understanding and addressing personality traits in healthcare professionals to promote their well-being and prevent this serious emotional, mental, and physical exhaustion state.

Opportunities and barriers for telemedicine in pain management: insights from a SIAARTI survey among Italian pain physicians.

BACKGROUND: The integration of telemedicine in pain management represents a significant advancement in healthcare delivery, offering opportunities to enhance patient access to specialized care, improve satisfaction, and streamline chronic pain management. Despite its growing adoption, there remains a lack of comprehensive data on its utilization in pain therapy, necessitating a deeper understanding of physicians' perspectives, experiences, and challenges. METHODS: A survey was conducted in Italy between January 2024 and May 2024. Specialist center members of the SIAARTI were sent an online questionnaire testing the state of the art of telemedicine for pain medicine. RESULTS: One-hundred thirty-one centers across Italy reveal varied adoption rates, with 40% routinely using telemedicine. Regional disparities exist, with Northern Italy showing higher adoption rates. Barriers include the absence of protocols, resource constraints, and bureaucratic obstacles. Despite challenges, telemedicine has shown positive impacts on service delivery, with increased service volume reported. Technological capabilities, including image sharing and teleconsultation with specialists, indicate promising interdisciplinary potential. CONCLUSIONS: The integration of advanced telemedicine software utilizing artificial intelligence holds promise for enhancing telemonitoring and alert systems, potentially leading to more proactive and personalized pain management strategies.

Data-driven Machine Learning Models for Risk Stratification and Prediction of Emergence Delirium in Pediatric Patients Underwent Tonsillectomy/Adenotonsillectomy.

AIM: In the pediatric surgical population, Emergence Delirium (ED) poses a significant challenge. This study aims to develop and validate machine learning (ML) models to identify key features associated with ED and predict its occurrence in children undergoing tonsillectomy or adenotonsillectomy. METHODS: The analysis involved data cleaning, exploratory data analysis (EDA), supervised predictive modeling, and unsupervised learning on a medical dataset (n = 423). After preliminary data cleaning, EDA encompassed plotting histograms, boxplots, pairplots, and correlation heatmaps to understand variable distributions and relationships. Four predictive models were trained including logistic regression (LR), random forest (RF), Support Vector Machine (SVM), and Gradient Boosting (XGBoost). The models were evaluated and compared using Receiver Operating Characteristic (ROC) Area Under the Curve (AUC), precision, recall, and feature importance. The RF model showed better performance and was used for the test (AUC-ROC 0.96, precision 1.00, and recall 0.92 on the validation set). K-means clustering was applied to find groups within the data. Elbow method and silhouette scores were used to determine the optimal number of clusters. The formed clusters were analyzed by aggregating features to understand the characteristics of each cluster. RESULTS: EDA revealed significant positive correlations between age, weight, American Society of Anesthesiologists (ASA) health score, and surgery duration with the risk of developing ED. Among the ML models, RF achieved the highest performance. Key predictive variables, based on the model's feature importance, included delirium screening scales, extubation time, and time to regain consciousness. Unsupervised K-means clustering identified 2-3 optimal clusters, which represented distinct patient subgroups: younger, healthier, low-risk individuals (cluster 0), and older patients with increasing chronic disease burden, higher delirium screening scores, and consequently higher post-operative delirium risk (clusters 1 and 2). CONCLUSIONS: ML techniques are valuable tools for extracting insights and making accurate predictions from healthcare data. High-performing algorithm-based models can be implemented for clinical decision support systems, facilitating early identification and intervention for ED in pediatric patients. By investigating various variables, it is possible to assess risk and implement preventive measures effectively. Furthermore, unsupervised clustering reveals distinct patient subgroups, enabling personalized perioperative management strategies and enhancing overall patient care.

Lung ultrasound among Expert operator'S: ScOring and iNter-rater reliability analysis (LESSON study) a secondary COWS study analysis from ITALUS group.

BACKGROUND: Lung ultrasonography (LUS) is a non-invasive imaging method used to diagnose and monitor conditions such as pulmonary edema, pneumonia, and pneumothorax. It is precious where other imaging techniques like CT scan or chest X-rays are of limited access, especially in low- and middle-income countries with reduced resources. Furthermore, LUS reduces radiation exposure and its related blood cancer adverse events, which is particularly relevant in children and young subjects. The score obtained with LUS allows semi-quantification of regional loss of aeration, and it can provide a valuable and reliable assessment of the severity of most respiratory diseases. However, inter-observer reliability of the score has never been systematically assessed. This study aims to assess experienced LUS operators' agreement on a sample of video clips showing predefined findings. METHODS: Twenty-five anonymized video clips comprehensively depicting the different values of LUS score were shown to renowned LUS experts blinded to patients' clinical data and the study's aims using an online form. Clips were acquired from five different ultrasound machines. Fleiss-Cohen weighted kappa was used to evaluate experts' agreement. RESULTS: Over a period of 3 months, 20 experienced operators completed the assessment. Most worked in the ICU (10), ED (6), HDU (2), cardiology ward (1), or obstetric/gynecology department (1). The proportional LUS score mean was 15.3 (SD 1.6). Inter-rater agreement varied: 6 clips had full agreement, 3 had 19 out of 20 raters agreeing, and 3 had 18 agreeing, while the remaining 13 had 17 or fewer people agreeing on the assigned score. Scores 0 and score 3 were more reproducible than scores 1 and 2. Fleiss' Kappa for overall answers was 0.87 (95% CI 0.815-0.931, p < 0.001). CONCLUSIONS: The inter-rater agreement between experienced LUS operators is very high, although not perfect. The strong agreement and the small variance enable us to say that a 20% tolerance around a measured value of a LUS score is a reliable estimate of the patient's true LUS score, resulting in reduced variability in score interpretation and greater confidence in its clinical use.

Continuous vs intermittent Non-Invasive blood pressure MONitoring in preventing postoperative organ failure (niMON): study protocol for an open-label, multicenter randomized trial.

BACKGROUND: Blood pressure has become one of the most important vital signs to monitor in the perioperative setting. Recently, the Italian Society of Anesthesia Analgesia Resuscitation and Intensive Care (SIAARTI) recommended, with low level of evidence, continuous monitoring of blood pressure during the intraoperative period. Continuous monitoring allows for early detection of hypotension, which may potentially lead to a timely treatment. Whether the ability to detect more hypotension events by continuous noninvasive blood pressure (C-NiBP) monitoring can improve patient outcomes is still unclear. Here, we report the rationale, study design, and statistical analysis plan of the niMON trial, which aims to evaluate the effect of intraoperative C-NiBP compared with intermittent (I-NiBP) monitoring on postoperative myocardial and renal injury. METHODS: The niMon trial is an investigator-initiated, multicenter, international, open-label, parallel-group, randomized clinical trial. Eligible patients will be randomized in a 1:1 ratio to receive C-NiBP or I-NiBP as an intraoperative monitoring strategy. The proportion of patients who develop myocardial injury in the first postoperative week is the primary outcome; the secondary outcomes are the proportions of patients who develop postoperative AKI, in-hospital mortality rate, and 30 and 90 postoperative days events. A sample size of 1265 patients will provide a power of 80% to detect a 4% absolute reduction in the rate of the primary outcome. CONCLUSIONS: The niMON data will provide evidence to guide the choice of the most appropriate intraoperative blood pressure monitoring strategy. CLINICAL TRIAL REGISTRATION: Clinical Trial Registration: NCT05496322, registered on the 5th of August 2023.

Effect on post-operative pulmonary complications frequency of high flow nasal oxygen versus standard oxygen therapy in patients undergoing esophagectomy for cancer: study protocol for a randomized controlled trial-OSSIGENA study.

Background: Postoperative pulmonary complications (PPCs) remain a challenge after esophagectomy. Despite improvement in surgical and anesthesiological management, PPCs are reported in as many as 40% of patients. The main aim of this study is to investigate whether early application of high-flow nasal cannula (HFNC) after extubation will provide benefit in terms of reduced PPC frequency compared to standard oxygen therapy. Methods: Patients aged 18-85 years undergoing esophagectomy for cancer treatment with radical intent, excluding those with American Society of Anesthesiologists (ASA) score >3 and severe systemic comorbidity (cardiac, pulmonary, renal or hepatic disease) will be randomized at the end of surgery to receive HFNC or standard oxygen therapy (Venturi mask or nasal goggles) after early extubation (within 12 hours after the end of surgery) for 48 hours. The main postoperative goals are to obtain SpO2 ≥94% and adequate pain control. Oxygen therapy after 48 hours will be stopped unless the physician deems it necessary. In case of respiratory clinical worsening, patients will be supported with the most appropriate tool (noninvasive ventilation or invasive mechanical ventilation). Pulmonary [pneumonia, pleural effusion, pneumothorax, atelectasis, acute respiratory distress syndrome (ARDS), tracheo-bronchial injury, air leak, reintubation, and/or respiratory failure] complications will be recorded as main outcome. Secondary outcomes, including cardiovascular, surgical, renal and infective complications will also be recorded. The primary analysis will be carried out on 320 patients (160 per group) and performed on an intention-to-treat (ITT) basis, including all participants randomized into the treatment groups, regardless of protocol adherence. The primary outcome, the PPC rate, will be compared between the two treatment groups using a chi-square test for categorical data, or Fisher's exact test will be used if the assumptions for the chi-square test are not met. Discussion: Recent evidence demonstrated that early application of HFNC improved the respiratory rate oxygenation index (ROX index) after esophagectomy but did not reduce PPCs. This randomized controlled multicenter trial aims to assess the potential effect of the application of HFNC versus standard oxygen over PPCs in patients undergoing esophagectomy. Trial Registration: This study is registered at clinicaltrial.gov NCT05718284, dated 30 January 2023.

Accuracy of Preoperative Lung Ultrasound Score for the prediction of Major Adverse Cardiac Events in elderly patients undergoing HIP Surgery under Spinal Anesthesia: the LUSHIP multicenter observational prospective study.

BACKGROUND AND OBJECTIVE: We hypothesize that lung ultrasound scores (LUS) can help stratify the cardiac risk of elderly patients undergoing orthopedic surgery for hip fracture, adding value to the Revised Cardiac Risk Index (RCRI), the American Society of Anesthesiologists Physical Status (ASA-PS) and the National Surgical Quality Improvement Program Myocardial infarction and Cardiac arrest (NSQIP-MICA). METHODS: Prospective, observational multicenter study of 11 Italian hospitals on patients aged >65 years with hip fractures needing urgent surgery. Subjects with major adverse cardiovascular events (MACE) in the previous 6 months or with ongoing acute heart failure were excluded. Trained anesthesiologists obtained preoperative LUS scores during preoperative evaluation. ROC curve analysis and comparison were used to evaluate test accuracy. RESULTS: A total of 877 patients were enrolled in the study period. 108 MACE events occurred in 98 patients, with an overall incidence of 11.2%. LUS score was higher in complicated than non-complicated patients, 11.6 ± 6.64 vs. 4.97 ± 4.90 (p < 0.001). Preoperative LUS score ≥8 showed both better AUC (0.78) and accuracy (0.76) in predicting MACE than the RCRI scores (p < 0.001), MICA scores (p = 0.001) and ASA classes (p < 0.001). LUS sensitivity was 0.71, specificity was 0.76, negative predictive value was 0.95. LUS score ≥8 showed an OR for MACE of 5.81[95% CI 3.55-9.69] at multivariate analysis. 91 patients (10.4%) experienced postoperative pneumonia showing a preoperative LUS score higher in the non-pneumonia group, p < 0.001. CONCLUSIONS: The preoperative LUS score, with its high negative predictive value, could improve patients' risk stratification when used alone or add further value to the RCRI score. REGISTRATION: Registered at clinicaltrials.gov as NCT04074876.

Supraspinal neuroinflammation and anxio-depressive-like behaviors in young- and older- adult mice with osteoarthritis pain: the effect of morphine.

RATIONALE: Asteoarthritis (OA) is a leading cause of chronic pain in the elderly population and is often associated with emotional comorbidities such as anxiety and depression. Despite age is a risk factor for both OA and mood disorders, preclinical studies are mainly conducted in young adult animals. OBJECTIVES: Here, using young adult (11-week-old) and older adult (20-month-old) mice, we evaluate in a monosodium-iodoacetate-(MIA)-induced OA model the development of anxio-depressive-like behaviors and whether brain neuroinflammation may underlie the observed changes. We also test whether an effective pain treatment may prevent behavioral and biochemical alterations. METHODS: Mechanical allodynia was monitored throughout the experimental protocol, while at the end of protocol (14 days), anxio-depressive-like behaviors and cognitive dysfunction were assessed. Neuroinflammatory condition was evaluated in prefrontal cortex, hippocampus and hypothalamus. Serum IFNγ levels were also measured. Moreover, we test the efficacy of a 1-week treatment with morphine (2.5 mg/kg) on pain, mood alterations and neuroinflammation. RESULTS: We observed that young adult and older adult controls (CTRs) mice had comparable allodynic thresholds and developed similar allodynia after MIA injection. Older adult CTRs were characterized by altered behavior in the tests used to assess the presence of depression and cognitive impairment and by elevated neuroinflammatory markers in brain areas compared to younger ones. The presence of pain induced depressive-like behavior and neuroinflammation in adult young mice, anxiety-like behavior in both age groups and worsened neuroinflammation in older adult mice. Morphine treatment counteracted pain, anxio-depressive behaviors and neuroinflammatory activation in both young adult and older adult mice. CONCLUSIONS: Here, we demonstrated that the presence of chronic pain in young adult mice induces mood alterations and supraspinal biochemical changes and aggravates the alterations already evident in older adult animals. A treatment with morphine, counteracting the pain, prevents the development of anxio-depressive disorders and reduces neuroinflammation.

Wireless Peripheral Nerve Stimulation for The Upper Limb: A Case Report.

Peripheral neuro-stimulation (PNS) has been proved to be effective for the treatment of neuropathic pain as well as other painful conditions. We discuss two approaches to PNS placement in the upper extremity. The first case describes a neuropathic syndrome after the traumatic amputation of the distal phalanx of the fifth digit secondary to a work accident with lack of responsiveness to a triple conservative therapy. An upper arm region approach for the PNS was chosen. The procedure had a favorable outcome; in fact, after one month the pain symptoms were absent (VAS 0) and the pharmacological therapy was suspended. The second case presented a patient affected by progressive CRPS type II in the sensory regions of the ulnar and median nerve in the hand, unresponsive to drug therapy. For this procedure, the PNS device was implanted in the forearm. Unfortunately, in this second case the migration of the catheter affected the effectiveness of the treatment. After examining the two cases in this paper, we changed our practice and suggest the implantation of PNS for radial, median and/or ulnar nerve stimulation in the upper arm region, which has significant advantages over the forearm region.

The Clinical Researcher Journey in the Artificial Intelligence Era: The PAC-MAN's Challenge.

Artificial intelligence (AI) is a powerful tool that can assist researchers and clinicians in various settings. However, like any technology, it must be used with caution and awareness as there are numerous potential pitfalls. To provide a creative analogy, we have likened research to the PAC-MAN classic arcade video game. Just as the protagonist of the game is constantly seeking data, researchers are constantly seeking information that must be acquired and managed within the constraints of the research rules. In our analogy, the obstacles that researchers face are represented by "ghosts", which symbolize major ethical concerns, low-quality data, legal issues, and educational challenges. In short, clinical researchers need to meticulously collect and analyze data from various sources, often navigating through intricate and nuanced challenges to ensure that the data they obtain are both precise and pertinent to their research inquiry. Reflecting on this analogy can foster a deeper comprehension of the significance of employing AI and other powerful technologies with heightened awareness and attentiveness.

Artificial Intelligence for Automatic Pain Assessment: Research Methods and Perspectives.

Although proper pain evaluation is mandatory for establishing the appropriate therapy, self-reported pain level assessment has several limitations. Data-driven artificial intelligence (AI) methods can be employed for research on automatic pain assessment (APA). The goal is the development of objective, standardized, and generalizable instruments useful for pain assessment in different clinical contexts. The purpose of this article is to discuss the state of the art of research and perspectives on APA applications in both research and clinical scenarios. Principles of AI functioning will be addressed. For narrative purposes, AI-based methods are grouped into behavioral-based approaches and neurophysiology-based pain detection methods. Since pain is generally accompanied by spontaneous facial behaviors, several approaches for APA are based on image classification and feature extraction. Language features through natural language strategies, body postures, and respiratory-derived elements are other investigated behavioral-based approaches. Neurophysiology-based pain detection is obtained through electroencephalography, electromyography, electrodermal activity, and other biosignals. Recent approaches involve multimode strategies by combining behaviors with neurophysiological findings. Concerning methods, early studies were conducted by machine learning algorithms such as support vector machine, decision tree, and random forest classifiers. More recently, artificial neural networks such as convolutional and recurrent neural network algorithms are implemented, even in combination. Collaboration programs involving clinicians and computer scientists must be aimed at structuring and processing robust datasets that can be used in various settings, from acute to different chronic pain conditions. Finally, it is crucial to apply the concepts of explainability and ethics when examining AI applications for pain research and management.