Senior Physician Research Articles

Deep learning based clinical target volumes contouring for prostate cancer: Easy and efficient application.

Radiotherapy has been crucial in prostate cancer treatment. However, manual segmentation is labor intensive and highly variable among radiation oncologists. In this study, a deep learning based automated contouring model is constructed for clinical target volumes (CTVs) of intact and postoperative prostate cancer. Computed tomography (CT) data sets of 197 prostate cancer patients were collected. Two auto-delineation models were built for radical radiotherapy and postoperative radiotherapy of prostate cancer respectively, and each model included CTVn for pelvic lymph nodes and CTVp for prostate tumors or prostate tumor beds. In the radical radiotherapy model, the volumetric dice (VD) coefficient of CTVn calculated by AI, was higher than that of the one delineated by the junior physicians (0.85vs. 0.82, p=0.018); In the postoperative radiotherapy model, the quantitative parameter of CTVn and CTVp, counted by AI, was better than that of the junior physicians. The median delineation time for AI was 0.23min in the postoperative model and 0.26min in the radical model, which were significantly shorter than those of the physicians (50.40 and 45.43min, respectively, p<0.001). The correction time of the senior physician for AI was much shorter compared with that for the junior physicians in both models (p<0.001). Using deep learning and attention mechanism, a highly consistent and time-saving contouring model was built for CTVs of pelvic lymph nodes and prostate tumors or prostate tumor beds for prostate cancer, which also might be a good approach to train junior radiation oncologists.

Preparing porcine lens to mimic human lens capsule.

To develop a chemical method that makes porcine lens anterior capsule resemble human lens anterior capsule in tear force and perforating force. Beijing Tongren Hospital, Beijing, China. Experimental study. Porcine eyes were divided into groups, and reagents (0.9% physiological saline, 0.1% sodium hypochlorite, 0.3% sodium hypochlorite, and 0.5% sodium hypochlorite) were injected into the anterior chamber, respectively, recorded as Groups A, B, C, and D, respectively. A senior physician collected each group of anterior capsules after performing continuous circular capsulorhexis and assessing the anterior capsule's tearing and perforation forces. An additional group, which consisted of human lens anterior capsules taken in the operating room from patients with cataract, recorded as Group E. A tensile system was used to measure each sample's tensile force. A significant difference was found between Group A and any other group in maximum tensile force and average tensile force in both transverse and longitudinal directions. No significant difference was found between any 2 groups from Group B to Group E. According to the surgeon's assessment, the tear force characteristic of the porcine lens anterior capsule treated with 0.1%, 0.3%, and 0.5% sodium hypochlorite solution was similar to that of the human lens anterior capsule. Porcine lens capsule treated using this method can be used for training of new surgeons. The porcine lens anterior capsule treated with 0.5% sodium hypochlorite, which results showed most resembled human lens anterior capsule, can be used for robotic training.

Anxiety in Endometriosis Patients: Implications for Clinical Practice

AbstractEndometriosis is a chronic disease associated with pain that affects at least 10% of all women of reproductive age. The symptoms of endometriosis have a negative impact on quality of life; they limit the patient physically and psychologically and are the cause of significant anxiety. The aim of our study was to investigate the anxiety levels of women presenting to our specialized endometriosis outpatient clinic at the RWTH Aachen University Hospital. We sought to record their anxiety symptoms and investigate the effect of extensive counseling on anxiety in patients with endometriosis. We hypothesized that detailed counseling and the planning of multimodal endometriosis therapy would reduce anxiety.Data from 183 patients with lower abdominal pain or suspected endometriosis were analyzed. Prior to the examination, the patients completed a standardized anamnesis form including the German version of the STAI (State Trait Anxiety Inventory). The diagnosis of endometriosis was made in accordance with the recent ESHRE guideline and was based on anamnesis, gynecological examination, and ultrasound by a specialist senior physician with many years of experience in treating endometriosis.Prior to the medical examination, we observed a mean STAI-State score of 48.44 ± 11.56 and STAI-Trait score 45.68 ± 11.27. Following medical examination and planning of multimodal therapy, there was a significant decrease in the mean STAI-State score from 48.44 ± 11.56 to 42.43 ± 10.01 (p &lt; 0.0001) and a minimal but significant decrease in the mean STAI-Trait score from 45.68 ± 11.27 to 45.01 ± 11.57 (p &lt; 0.05).Our study showed that endometriosis is associated with high levels of anxiety. Detailed counseling and the planning of a multimodal therapy led to a significant improvement in anxiety levels.

Deep learning radiomics based on ultrasound images for the assisted diagnosis of chronic kidney disease.

This study aimed to explore the value of ultrasound (US) images in chronic kidney disease (CKD) screening by constructing a CKD screening model based on grey-scale US images. According to the CKD diagnostic criteria, 1049 patients from Tongde Hospital of Zhejiang Province were retrospectively enrolled in the study. A total of 4365 renal US images were collected from these patients. Convolutional neural networks were used for feature extractions and a screening model was constructed by fusing ResNet34 and texture features to identify CKD and its stage. A comparative analysis was performed to compare the diagnosis results of the model with physicians. When diagnosing CKD or non-CKD, the receiver operating characteristic curve (AUC) of our model was 0.918 and that of the senior physician group was 0.869 (p < .05). For the diagnosis of CKD stage, the AUC of our model for CKD G1-G3 was 0.781, 0.880, and 0.905, respectively, while the AUC of the senior physician group for CKD G1-G3 was 0.506, 0.586, and 0.796, respectively; all differences were statistically significant (p < .05). The diagnostic efficiency of our model for CKD G4 and G5 reached the level of the senior physicians group. Specifically, the AUC of our model for CKD G4-G5 was 0.867 and 0.931, respectively, while the AUC of the senior physician group for CKD G4-G5 was 0.838 and 0.963, respectively (all p > .05). Our deep learning radiomics model is more effective than senior physicians in the diagnosis of early CKD.

Comparison of Medical Research Abstracts Written by Surgical Trainees and Senior Surgeons or Generated by Large Language Models

Artificial intelligence (AI) has permeated academia, especially OpenAI Chat Generative Pretrained Transformer (ChatGPT), a large language model. However, little has been reported on its use in medical research. To assess a chatbot's capability to generate and grade medical research abstracts. In this cross-sectional study, ChatGPT versions 3.5 and 4.0 (referred to as chatbot 1 and chatbot 2) were coached to generate 10 abstracts by providing background literature, prompts, analyzed data for each topic, and 10 previously presented, unassociated abstracts to serve as models. The study was conducted between August 2023 and February 2024 (including data analysis). Abstract versions utilizing the same topic and data were written by a surgical trainee or a senior physician or generated by chatbot 1 and chatbot 2 for comparison. The 10 training abstracts were written by 8 surgical residents or fellows, edited by the same senior surgeon, at a high-volume hospital in the Southeastern US with an emphasis on outcomes-based research. Abstract comparison was then based on 10 abstracts written by 5 surgical trainees within the first 6 months of their research year, edited by the same senior author. The primary outcome measurements were the abstract grades using 10- and 20-point scales and ranks (first to fourth). Abstract versions by chatbot 1, chatbot 2, junior residents, and the senior author were compared and judged by blinded surgeon-reviewers as well as both chatbot models. Five academic attending surgeons from Denmark, the UK, and the US, with extensive experience in surgical organizations, research, and abstract evaluation served as reviewers. Surgeon-reviewers were unable to differentiate between abstract versions. Each reviewer ranked an AI-generated version first at least once. Abstracts demonstrated no difference in their median (IQR) 10-point scores (resident, 7.0 [6.0-8.0]; senior author, 7.0 [6.0-8.0]; chatbot 1, 7.0 [6.0-8.0]; chatbot 2, 7.0 [6.0-8.0]; P = .61), 20-point scores (resident, 14.0 [12.0-7.0]; senior author, 15.0 [13.0-17.0]; chatbot 1, 14.0 [12.0-16.0]; chatbot 2, 14.0 [13.0-16.0]; P = .50), or rank (resident, 3.0 [1.0-4.0]; senior author, 2.0 [1.0-4.0]; chatbot 1, 3.0 [2.0-4.0]; chatbot 2, 2.0 [1.0-3.0]; P = .14). The abstract grades given by chatbot 1 were comparable to the surgeon-reviewers' grades. However, chatbot 2 graded more favorably than the surgeon-reviewers and chatbot 1. Median (IQR) chatbot 2-reviewer grades were higher than surgeon-reviewer grades of all 4 abstract versions (resident, 14.0 [12.0-17.0] vs 16.9 [16.0-17.5]; P = .02; senior author, 15.0 [13.0-17.0] vs 17.0 [16.5-18.0]; P = .03; chatbot 1, 14.0 [12.0-16.0] vs 17.8 [17.5-18.5]; P = .002; chatbot 2, 14.0 [13.0-16.0] vs 16.8 [14.5-18.0]; P = .04). When comparing the grades of the 2 chatbots, chatbot 2 gave higher median (IQR) grades for abstracts than chatbot 1 (resident, 14.0 [13.0-15.0] vs 16.9 [16.0-17.5]; P = .003; senior author, 13.5 [13.0-15.5] vs 17.0 [16.5-18.0]; P = .004; chatbot 1, 14.5 [13.0-15.0] vs 17.8 [17.5-18.5]; P = .003; chatbot 2, 14.0 [13.0-15.0] vs 16.8 [14.5-18.0]; P = .01). In this cross-sectional study, trained chatbots generated convincing medical abstracts, undifferentiable from resident or senior author drafts. Chatbot 1 graded abstracts similarly to surgeon-reviewers, while chatbot 2 was less stringent. These findings may assist surgeon-scientists in successfully implementing AI in medical research.

Advanced Practice Provider Professional Advancement Model: A 10-Year Experience

In a large 700-bed academic medical center, a professional advancement model for advanced practice providers (APPs) provides a structured process to guide professional growth, development, and promotion. During a 10-year period, 217 APPs applied and 162 were promoted, resulting in a 74% promotion rate. Of the APPs promoted to senior nurse practitioner or senior physician assistant, 140 (86%) are still employed at the medical center. This advancement model provides a structured process to guide the APP workforce in developing clinical expertise and fostering professional development while improving retention and overall satisfaction.

Ongoing decision-making dilemma for treatment of de novo spinal infections: a comparison of the Spinal Infection Treatment Evaluation Score with the Spinal Instability Spondylodiscitis Score and Spine Instability Neoplastic Score.

De novo spinal infections are an increasing medical problem. The decision-making for surgical or nonsurgical treatment for de novo spinal infections is often a non-evidence-based process and commonly a case-by-case decision by single physicians. A scoring system based on the latest evidence might help improve the decision-making process compared with other purely radiology-based scoring systems or the judgment of a single senior physician. Patients older than 18 years with an infection of the spine who underwent nonsurgical or surgical treatment between 2019 and 2021 were identified. Clinical data for neurological status, pain, and existing comorbidities were gathered and transferred to an anonymous spreadsheet. Patients without an MR image and a CT scan of the affected spine region were excluded from the investigation. A multidisciplinary expert panel used the Spine Instability Neoplastic Score (SINS), Spinal Instability Spondylodiscitis Score (SISS), and Spinal Infection Treatment Evaluation Score (SITE Score), previously developed by the authors' group, on every clinical case. Each physician of the expert panel gave an individual treatment recommendation for surgical or nonsurgical treatment for each patient. Treatment recommendations formed the expert panel opinion, which was used to calculate predictive validities for each score. A total of 263 patients with spinal infections were identified. After the exclusion of doubled patients, patients without de novo infections, or those without CT and MRI scans, 123 patients remained for the investigation. Overall, 70.70% of patients were treated surgically and 29.30% were treated nonoperatively. Intraclass correlation coefficients (ICCs) for the SITE Score, SINS, and SISS were 0.94 (95% CI 0.91-0.95, p < 0.01), 0.65 (95% CI 0.91-0.83, p < 0.01), and 0.80 (95% CI 0.91-0.89, p < 0.01). In comparison with the expert panel decision, the SITE Score reached a sensitivity of 96.97% and a specificity of 81.90% for all included patients. For potentially unstable and unstable lesions, the SISS and the SINS yielded sensitivities of 84.42% and 64.07%, respectively, and specificities of 31.16% and 56.52%, respectively. The SITE Score showed higher overall sensitivity with 97.53% and a higher specificity for patients with epidural abscesses (75.00%) compared with potentially unstable and unstable lesions for the SINS and the SISS. The SITE Score showed a significantly higher agreement for the definitive treatment decision regarding the expert panel decision, compared with the decision by a single physician for patients with spondylodiscitis, discitis, or spinal osteomyelitis. The SITE Score shows high sensitivity and specificity regarding the treatment recommendation by a multidisciplinary expert panel. The SITE Score shows higher predictive validity compared with radiology-based scoring systems or a single physician and demonstrates a high validity for patients with epidural abscesses.

The Impact of Involving a Senior Emergency Physician in the Triage Process

BACKGROUND: Emergency Department (ED) overcrowding has been demonstrated in several studies to be associated with undesirable effects such as longer waiting times, reduced patient satisfaction, and, most importantly, poor patient outcomes. Furthermore, long waiting times for walk-ins result in more complaints and patient dissatisfaction than illness management itself, with the majority of issues arising as a result of real and perceived waiting periods before being seen by the doctor. AIM: We set out to investigate whether introducing a senior emergency physician into the triage system would reduce waiting time, door-to-decision time, and door-to-doctor time, as well as increase patient satisfaction across the ED. METHOD: This was an interventional pre-post study that utilised retrospective data to evaluate the effect on ED throughput of triage by senior emergency physicians. We aimed to measure its impact on waiting time, door-to-decision time, and door-to-doctor time, along with ED patient satisfaction. RESULTS: Patient satisfaction, the overall assessment of treatment received during the visit, increased, from 74.975 to 77.425, and the likelihood of patients recommending the ED increased from 71.36 to 75.21. Operational metrics revealed a considerable drop in door-to-decision time (admit or discharge) of 46 minutes and 3 seconds, as well as a 1 minute and 21 second reduction in time from door to doctor (arrival to first provider). CONCLUSION: The mixed results hint at an effective but iterative process of enhancing patient flow and experience in the ED through senior physician triage.

Randomized controlled open-label trial to evaluate prioritization software for the secondary triage of patients in the pediatric emergency department

BackgroundThe continual increase in patient attendance at the emergency department (ED) is a worldwide health issue. The aim of this study was to determine whether the use of a secondary prioritization software reduces the patients’ median length of stay (LOS) in the pediatric ED.MethodsA randomized, controlled, open-label trial was conducted over a 30-day period between March 15th and April 23rd 2021 at Lille University Hospital. Work days were randomized to use the patient prioritization software or the pediatric ED’s standard dashboard. All time intervals between admission and discharge were recorded prospectively by a physician not involved in patient care during the study period. The study’s primary endpoint was the LOS in the pediatric ED, which was expected to be 15 min shorter in the intervention group than in the control group. The secondary endpoints were specific time intervals during the stay in the pediatric ED and levels of staff satisfaction.Results1599 patients were included: 798 in the intervention group and 801 in the control group. The median [interquartile range] LOS was 172 min [113–255] in the intervention group and 167 min [108–254) in the control group (p = 0.46). In the intervention group, the time interval between admission to the first medical evaluation for high-priority patients and the time interval between the senior physician’s final evaluation and patient discharge were shorter (p < 0.01). The median satisfaction score was 68 [55–80] (average).ConclusionThe patients’ total LOS was not significantly shorter on days of intervention. However, use of the electronic patient prioritization tool was associated with significant decreases in some important time intervals during care in the pediatric ED.ClinicalTrials.gov: NCT05994196Trial registration number: NCT05994196. Date of registration: August 16th, 2023

Effectiveness of nirmatrelvir-ritonavir versus azvudine for adult inpatients with severe or critical COVID-19

BackgroundIn China, both nirmatrelvir-ritonavir (Paxlovid) and azvudine have been granted approval to treat adult SARS-CoV-2-infected patients with moderate symptoms. Information about the clinical effect of the two available agents among...

Complications Rate and a Multidimensional Analysis of Their Causes of Tube Thoracostomy: A Mixed-Methods Study.

Tube thoracostomy (TT) complications are common in respiratory medicine. However, the prevalence of complications and risk factors is unknown, and data on countermeasures are lacking. This was a mixed-methods retrospective observational and qualitative study. This retrospective observational study included TT performed on patients admitted to the Department of Respiratory Medicine at our University Hospital between January 1, 2019, and August 31, 2022 (n=169). The primary endpoint was the incidence of TT-related complications. We reviewed the association between complications and patient- and medical-related factors as secondary endpoints. In this qualitative study, we theorized the background of physicians' susceptibility to TT-related complications based on the grounded theory approach. Complications were observed in 20 (11.8%) of the 169 procedures; however, they were unrelated to 30-day mortality. Poor activities of daily living (odds ratio 4.3, p=0.007) and regular administration of oral steroids (odds ratio 3.1, p=0.025) were identified as patient-related risk factors. Physicians undergoing training caused the most complications, and the absence of a senior physician at the procedure site (odds ratio 3.5, p=0.031) was identified as a medical risk factor. Based on this qualitative study, we developed a new model for TT complication rates consistent with the relationship between physicians' professional skills, professional identity, and work environments. Complications associated with TT are common. Therefore, it is necessary to implement measures similar to those identified in this study. Particularly, a supportive environment should be established for the training of physicians.

Abstract 3569: Using AI to automatically process data from unstructured health records of patients with lung cancer

Abstract Introduction: Structured databases created from electronic health records (EHR) are crucial for cancer research. Manual data entry into databases is both labor-intensive and error-prone. We aimed to develop an artificial intelligence (AI)-driven approach for automatically inputting patient information from EHRs. Methods: A REDCap database for 53 patients with advanced lung cancer treated with first line immuno+/-chemotherapy at Gustave Roussy between 02/2021-06/2023 was manually populated by physicians with demographics, risk factors, cancer history, and treatment data, with 137 variables/patient. Given unstructured medical letters and a schematic description of each variable, generative AI was used to find, quote and process variables into a structured form. We directed large language model actions with prompt engineering and tailored few-shot examples. Mortality data were auto-extracted from the French public registry, INSEE. We assessed consistency between manual (MDE) and automated data entry (ADE), with a secondary manual review for mismatches performed by a senior physician. Results: In total, 7,261 data points were assessed. ADE averaged 10 minutes per patient inclusive of quality checks vs 1 h for MDA. The concordance rate between ADE and MDE was 90.2% (6,550/7,261) with a discordance of 6.8% (496/7,261). Data was missing in 0.8% (59/7,261) of cases for both methods, 1% (73/7,261) for ADE and 1,1% (83/7,261) for MDA. After checking discordances, ADE correctness was 95.3% (6,922/7,261) (Table), MDE was 93.8% (6,809/7,261). Errors in ADE (207/7,261) were due to algorithm refinement needs (70%) and missing EHR information (30%). Conclusion: Generative AI has a strong potential for identifying and structuring data from EHRs, yielding consistency superior to manual entry by physicians and 85% reduced amount of time. This may enhance the efficiency, accuracy, and scalability of EHR-to-database conversions. ADE for &amp;gt;1000 patients will be presented at the meeting. ADE correctness 100% 95-99% 90-94% 80-89% Sex; Age; Life status; Last follow-up; Metastatic from diagnosis Molecular alterations; Metastatic sites; Progression sites; Treatment type Tobacco &amp; cannabis consumption; Histology; PDL1 score; Progression event Pack years; Date of diagnosis &amp; first metastasis; Stage; Date of start treatment Citation Format: Mihaela Aldea, Pierre Rolland, Solenne Simon, Aliette Poplu, Muriel Wartelle, Benjamin Vignal, Jean-Charles Louis, Francois Lion, Arnaud Borie, David Planchard, Caroline Robert, Stefan Michiels, Fabrice Andre, Fabrice Barlesi, Franck Le Ouay, Benjamin Besse. Using AI to automatically process data from unstructured health records of patients with lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3569.

Retrospective study on unilateral polyotia combined with microtia utilizing the technique of preserving residual ear tissue

The presence of polyotia in individuals with microtia is a rare deformity. Due to the intricate structure of the auricle, uncertain etiology, and challenging corrective techniques, it has always been a focal point in the field of plastic surgery. The present study presents a technique for correcting the combination of polyotia and microtia by utilizing residual ear tissue as graft material. The retrospective study included 23 patients with polyotia and microtia from 2018 to 2022. The residual ear tissue was used to rectify auricular deformities in all patients. The patients were instructed to evaluate the satisfaction of the auricle shape using a visual analog scale (VAS) both before and 6 months after the surgical procedure. The esthetic outcomes of auricle subunits were simultaneously assessed by a senior physician pre- and postoperatively. The mean duration of follow-up in this study was 8.73 months. The preoperative VAS satisfaction score was recorded as 2.26±0.86, while the post-operative VAS score significantly increased to 7.86±0.86. The preoperative auricle esthetic outcomes score was recorded as 9.95±1.74, while the post-operative score significantly increased to 24.04±2.16. The follow-up period did not present any cases of flap necrosis, hematoma, infection, or wound dehiscence. The study demonstrates that comprehensive utilization of residual auricular tissue can lead to optimal outcomes in correcting polyotia with concha-type microtia. The utilization of residual ear tissue can be maximized to streamline the operation, minimize bodily harm, and enhance patient satisfaction.

Use of a Handheld Ultrasonographic Device to Identify Heart Failure and Pulmonary Disease in Rural Africa

Agreement in lung ultrasonography findings between clinicians using a handheld ultrasonographic device and expert sonographers using a high-end ultrasonographic machine has not been studied in sub-Saharan Africa. To determine the agreement in ultrasonographic findings and diagnoses between primary care clinicians trained in lung ultrasonography, board-certified expert sonographers, and senior physicians. This cross-sectional single-center study was conducted from February 1, 2022, to April 30, 2023 at a referral center in rural Tanzania. Individuals 5 years or older with respiratory symptoms and at least 2 distinct respiratory signs or symptoms were eligible. A total of 459 individuals were screened. Participants provided their medical history and underwent a clinical examination and lung ultrasonography performed by a clinician, followed by a lung ultrasonography performed by an expert sonographer, and finally chest radiography and a final evaluation performed by a senior physician. Other tests, such as echocardiography and Mycobacterium tuberculosis testing, were conducted on the decision of the physician. Clinicians received 2 hours of instruction and three 2-hour sessions of clinical training in the use of a handheld lung ultrasonographic device; expert sonographers were board-certified. Percentage agreement and Cohen κ coefficient for sonographic findings and diagnoses compared between clinicians and expert sonographers, and between clinicians and senior physicians. The median (IQR) age of 438 included participants was 54 (38-66) years, and 225 (51%) were male. The median (range) percentage agreement of ultrasonographic findings between clinicians and expert sonographers was 93% (71%-99%), with κ ranging from -0.003 to 0.83. Median (range) agreement of diagnoses between clinicians and expert sonographers was 90% (50%-99%), with κ ranging from -0.002 to 0.76. Between clinicians and senior physicians, median (range) agreement of diagnoses was 89% (55%-90%), with κ ranging from -0.008 to 0.76. Between clinicians and senior physicians, diagnosis agreements were 85% (κ, 0.69) for heart failure, 78% (κ, 0.57) for definite or probable tuberculosis, 50% (κ, 0.002) for viral pneumonia, and 56% (κ, 0.06) for bacterial pneumonia. In this cross-sectional study, the agreement of ultrasonographic findings between clinicians and sonographers was mostly substantial. Between clinicians and senior physicians, agreement was substantial in the diagnosis of heart failure, moderate in the diagnosis of tuberculosis, but slight in the diagnosis of pneumonia. These findings suggest that handheld ultrasonographic devices used in addition to clinical examination may support clinicians in diagnosing cardiac and pulmonary diseases in rural sub-Saharan Africa.

Deep-learning-based renal artery stenosis diagnosis via multimodal fusion.

Diagnosing Renal artery stenosis (RAS) presents challenges. This research aimed to develop a deep learning model for the computer-aided diagnosis of RAS, utilizing multimodal fusion technology based on ultrasound scanning images, spectral waveforms, and clinical information. A total of 1485 patients received renal artery ultrasonography from Peking Union Medical College Hospital were included and their color doppler sonography (CDS) images were classified according to anatomical site and left-right orientation. The RAS diagnosis was modeled as a process involving feature extraction and multimodal fusion. Three deep learning (DL) models (ResNeSt, ResNet, and XCiT) were trained on a multimodal dataset consisted of CDS images, spectrum waveform images, and individual basic information. Predicted performance of different models were compared with senior physician and evaluated on a test dataset (N=117 patients) with renal artery angiography results. Sample sizes of training and validation datasets were 3292 and 169 respectively. On test data (N=676 samples), predicted accuracies of three DL models were more than 80% and the ResNeSt achieved the accuracy 83.49%±0.45%, precision 81.89%±3.00%, and recall 76.97%±3.7%. There was no significant difference between the accuracy of ResNeSt and ResNet (82.84%±1.52%), and the ResNeSt was higher than the XCiT (80.71%±2.23%, p<0.05). Compared to the gold standard, renal artery angiography, the accuracy of ResNest model was 78.25%±1.62%, which was inferior to the senior physician (90.09%). Besides, compared to the multimodal fusion model, the performance of single-modal model on spectrum waveform images was relatively lower. The DL multimodal fusion model shows promising results in assisting RAS diagnosis.

Heterogeneity in the role of emergency physicians and treatment of acute atrial fibrillation in emergency departments—results of the International Atrial Fibrillation Background (AFiB) Study

The consept of emergency departments (EDs) with specialized teams of emergency physicians originated in the United Kingdom and the United States during the 1970s and was expanded across most European countries in the twenty-first century. Among the various cardiac arrhythmias encountered in EDs, atrial fibrillation (AF) is the most prevalent, contributing to ED congestion. Existing guidelines offer multiple treatment options for acute-onset AF occurring within 48 hours. The aim of The Atrial Fibrillation Background Study is to evaluate treatment strategies, practices and the role of emergency physicians in managing acute-onset AF in Western medical tradition across Europe, the United States and China (Hong Kong). The data for this nonexperimental survey were collected through a questionnaire administered to the medical director or a senior physician at each of the 12 participating EDs. We obtained information regarding the total number of physicians employed in these EDs, their respective specialties, and the patient caseloads they managed. Additionally, we gathered data on the diagnostic and treatment protocols employed for atrial fibrillation (AF). In the investigated EDs in Hong Kong, the United Kingdom and the United States, patients were treated by emergency physicians. Comparatively, many European EDs primarily relied on physicians with traditional medical specialties. Diagnostic methods employed for acute AF ranged from point-of-care testing to comprehensive laboratory panels and echocardiography. In terms of AF treatment, rate control was the preferred approach in Hong Kong and the USA EDs, while rhythm control was preferred in European settings. Regarding rhythm control, there were considerable variations in preferences between pharmacological and electrical cardioversion methods. Findings from the AFiB Study highlight the growing significance of emergency physicians in the management of acute AF, as well as the divergent treatment approaches for acute AF observed in EDs between Europe and the United States.

Cost-effectiveness in an interprofessional training ward within a university department for internal medicine: a monocentric open-label controlled study of the A-STAR Regensburg.

Interprofessional collaboration in healthcare involves diverse professionals working together to address complex patient needs. Interprofessional training wards offer workplace-based interprofessional education in real healthcare settings, fostering collaborative learning among students. While their educational value is widely recognized, debates persist regarding their cost-effectiveness due to limited research. This study assesses the cost efficiency of the interprofessional training ward Regensburg (A-STAR) within the Department of Internal Medicine I at the University Hospital Regensburg, compared to conventional wards. From October 2019 to December 2022, 7,244 patient cases were assigned to A-STAR or conventional wards by case managers, with a comprehensive analysis of all associated revenues and costs. A-STAR treated 1,482 patients, whereas conventional wards treated 5,752 patients, with more males and younger patients at A-STAR. A-STAR achieved higher profit per case (€1,508.74) attributed to increased revenues and reduced material costs. It generated an average of €1,366.54 more Diagnosis Related Groups (DRG) revenue per case annually than conventional wards, due to greater medical complexity reflected in a higher case-mix index (CMI: 2.4 vs. 2.2). The increased case complexity led to longer patient stays (9.0 vs. 8.1 days) and fewer cases treated annually at A-STAR (27.4 cases/year vs. 37.8 cases/year). The higher CMI did not result in a higher proportion of patients requiring isolation. A-STAR exhibited a higher capacity utilization rate (87.1% vs. 83.9%). Personnel costs per case at A-STAR were initially elevated due to enhanced observation by the senior physician but were gradually mitigated by expanding A-STAR's bed capacity. Material costs were consistently lower on a per-case basis at A-STAR (€1512.02 vs. €1577.12), particularly in terms of medication expenses, indicating more resource-efficient operations. From the A-STAR graduates, 18 individuals were recruited for permanent positions as doctors or nurses over 2 years. A-STAR demonstrates economic efficiency and stability even during the COVID-19 pandemic. The substantial personnel acquisition is likely influenced by high levels of satisfaction with education and work and is economically relevant in medical staff shortages. These findings provide a compelling rationale for the broader implementation of interprofessional training wards, establishing them as vital platforms for nurturing future professionals.

Bridging AI and Clinical Practice: Integrating Automated Sleep Scoring Algorithm with Uncertainty-Guided Physician Review.

This study aims to enhance the clinical use of automated sleep-scoring algorithms by incorporating an uncertainty estimation approach to efficiently assist clinicians in the manual review of predicted hypnograms, a necessity due to the notable inter-scorer variability inherent in polysomnography (PSG) databases. Our efforts target the extent of review required to achieve predefined agreement levels, examining both in-domain (ID) and out-of-domain (OOD) data, and considering subjects' diagnoses. A total of 19,578 PSGs from 13 open-access databases were used to train U-Sleep, a state-of-the-art sleep-scoring algorithm. We leveraged a comprehensive clinical database of an additional 8832 PSGs, covering a full spectrum of ages (0-91 years) and sleep-disorders, to refine the U-Sleep, and to evaluate different uncertainty-quantification approaches, including our novel confidence network. The ID data consisted of PSGs scored by over 50 physicians, and the two OOD sets comprised recordings each scored by a unique senior physician. U-Sleep demonstrated robust performance, with Cohen's kappa (K) at 76.2% on ID and 73.8-78.8% on OOD data. The confidence network excelled at identifying uncertain predictions, achieving AUROC scores of 85.7% on ID and 82.5-85.6% on OOD data. Independently of sleep-disorder status, statistical evaluations revealed significant differences in confidence scores between aligning vs discording predictions, and significant correlations of confidence scores with classification performance metrics. To achieve κ ≥ 90% with physician intervention, examining less than 29.0% of uncertain epochs was required, substantially reducing physicians' workload, and facilitating near-perfect agreement. Inter-scorer variability limits the accuracy of the scoring algorithms to ~80%. By integrating an uncertainty estimation with U-Sleep, we enhance the review of predicted hypnograms, to align with the scoring taste of a responsible physician. Validated across ID and OOD data and various sleep-disorders, our approach offers a strategy to boost automated scoring tools' usability in clinical settings.

TRAINING IN HEALTH AND AGING POLICY MID CAREER—REFLECTING ON A POTENTIAL INFLECTION POINT

Abstract The Health and Aging Policy Fellowship (HAPF) trains fellows to understand policymaking with the goal to improve the lives of older people across life domains. Comparatively few mid-career or more senior experts in healthcare or aging have direct experience with federal or state policymaking, or working to inform evidence-based health policy. As a senior physician faculty member at a School of Medicine and School of Public Health, in recent years the author has worked at state and local levels to inform evidence-based health policy. The HAPF offered an opportunity for the author to improve knowledge and understanding of decision-making and policymaking in Washington DC and at the federal level. Fellowship-sponsored training improved fellows’ understanding of health policy and potential effectiveness in federal health policy environments. Trainings through HAPF, the American Political Science Association, and Academy Health provided greater breadth and depth of knowledge on roles of various agencies, executive and congressional policymaking, and understanding factors influencing policymaking. This training enhanced the author’s readiness for fellowship activities in a U.S. Senate office. Exposure to experts in health and aging opened the door to fellows’ participation in efforts to improve opportunities for healthy aging and to advance age-friendly health systems. Work in the U.S. Senate includes exposure to federal, state, non-governmental stakeholders who can work together to influence policy to improve the lives of aging and older people. The author will highlight specific examples of training, policy opportunities, and policy challenges in the context of his fellowship.

Application of Patient Sentiment Analysis to Evaluate Glaucoma Care

Patients utilize online physician reviews to decide between and rate ophthalmologists. Sentiment analysis allows for better understanding of patient experiences. In this study, Valence Aware Dictionary sEntiment Reasoner (VADER) and word frequency analysis of glaucoma specialist Healthgrades reviews were used to determine factors prioritized by patients. Retrospective cross-sectional analysis. N/A. Written reviews and Star ratings of glaucoma specialists listed under the Physicians Payments Sunshine Acts were obtained, and demographic information was collected. Valence Aware Dictionary sEntiment Reasoner produced Negative, Neutral, Positive, and Compound scores of reviews, and these were stratified by demographic variables. Word frequency review was applied to determine popular words and phrases. Star ratings, VADER Compound score of written reviews, and highest word frequencies. A total of 203 glaucoma specialists and 3531 written reviews were assessed. Glaucoma specialists had an average of 4.26/5 stars, with a mean of 30 ratings per physician on Healthgrades. Most physicians (86%) had overall Positive written reviews (VADER=0.74), indicating high patient satisfaction. Specialists who were women or had fewer years of practice had higher Compound and Star scores than their respective male and senior counterparts, with statistical significance observed between junior and senior physician Stars (P<0.001). Repeated words pertaining to the surgery, staff, wait times, and questions were common overall and among the most positive and most negative reviews. Glaucoma specialist patients value nonclinical factors, such as appointment setting and nonphysician health-care staff members, in their written reviews. Thus, factors beyond clinical outcomes are influential in the overall patient experience and should be considered to improve health-care delivery. These results can also advise ophthalmologists on factors that patients prioritize when evaluating physicians, which influences the decisions of other patients seeking glaucoma care. The author(s) have no proprietary or commercial interest in any materials discussed in this article.