Citations

Abstract Background: Many patients use artificial intelligence (AI) chatbots as a rapid source of health information. This raises important questions about the reliability and effectiveness of AI chatbots in delivering accurate and understandable information. Purpose: To evaluate and compare the accuracy, conciseness, and readability of responses from OpenAI ChatGPT-4 and Google Bard to patient inquiries concerning the novel 177Lu-PSMA-617 therapy for prostate cancer. Materials and Methods: Two experts listed the 12 most commonly asked questions by patients on 177Lu-PSMA-617 therapy. These twelve questions were prompted to OpenAI ChatGPT-4 and Google Bard. AI-generated responses were distributed using an online survey platform (Qualtrics) and blindly rated by eight experts. The performances of the AI chatbots were evaluated and compared across three domains: accuracy, conciseness, and readability. Additionally, potential safety concerns associated with AI-generated answers were also examined. The Mann-Whitney U and chi-square tests were utilized to compare the performances of AI chatbots. Results: Eight experts participated in the survey, evaluating 12 AI-generated responses across the three domains of accuracy, conciseness, and readability, resulting in 96 assessments (12 responses x 8 experts) for each domain per chatbot. ChatGPT-4 provided more accurate answers than Bard (2.95 ± 0.671 vs 2.73 ± 0.732, p=0.027). Bard's responses had better readability than ChatGPT-4 (2.79 ± 0.408 vs 2.94 ± 0.243, p=0.003). Both ChatGPT-4 and Bard achieved comparable conciseness scores (3.14 ± 0.659 vs 3.11 ± 0.679, p=0.798). Experts categorized the AI-generated responses as incorrect or partially correct at a rate of 16.6% for ChatGPT-4 and 29.1% for Bard. Bard's answers contained significantly more misleading information than those of ChatGPT-4 (p = 0.039). Conclusion: AI chatbots have gained significant attention, and their performance is continuously improving. Nonetheless, these technologies still need further improvements to be considered reliable and credible sources for patients seeking medical information on 177Lu-PSMA-617 therapy.

ObjectivesOur institution introduced universal vasa previa (VP) screening utilizing transabdominal ultrasound with color Doppler for all pregnancies at the second trimester anatomy scan. Our study sought to describe the clinical impact of this intervention.MethodsRadiology records from the 12 months pre‐ and post‐intervention were queried for “vasa previa.” Records included for analysis were those with a first‐time diagnosis or discussion of VP at the anatomy scan. Cases were categorized by outcome: (Group 1) True VP, with subgroups A, unresolved by time of delivery and B, resolved by delivery; (Group 2) False positives; (Group 3) Possible VP without definitive diagnosis; and (Group 4) VP ruled out, for example, “no features of VP.” Group size was expressed as a percentage of total anatomy scans during pre‐ or post‐intervention periods respectively. Absolute and relative percent change were calculated for each group.ResultsIn the pre‐intervention period, 1 case (0.36% of total scans) was categorized in Group 1A, 1 case (0.36%) in Group 3, and 7 cases (2.53%) in Group 4. In the post‐intervention period, 2 cases (0.30%) were in Group 1A, 4 cases (0.61%) in Group 1B, 2 cases (0.30%) in Group 2, 1 case (0.15%) in Group 3, and 7 cases (1.06%) in Group 4. There was a +153% relative change in true positives, from 0.36 to 0.91%.ConclusionsUniversal color Doppler screening may have increased detection (sensitivity) while simultaneously increasing false positives (decreased specificity). While decreasing sensitivity is not ideal, this is acceptable given the potential catastrophic outcome of a missed VP.

AbstractIntroduction/PurposePercutaneous core‐needle biopsy of the testicle has been shown to be a safe and effective method of obtaining tissue for histological analysis and can be considered in specific clinical scenarios. While the use of spermatic cord block has been shown to be effective in pain relief in the emergent setting and as an anaesthetic option for inguinal surgery, its use in percutaneous core‐needle biopsy has not been well described. Through this case series, we present our experience and technique of ultrasound‐guided percutaneous core‐needle biopsy using spermatic cord block in the setting of indeterminant testicular masses.MethodsOur departmental biopsy database was reviewed to identify patients who underwent percutaneous core‐needle biopsy of the testicle from March 2010 to July 2022 and who also received spermatic cord block during the procedure.ResultsThree patients were identified who met the search criteria. All three patients presented for the evaluation of indeterminant testicular mass and had a known non‐testicular primary cancer diagnosis at the time of biopsy. All three biopsies were performed using a combination of spermatic cord block, moderate sedation, and local anaesthetic. Biopsies were obtained using an 18‐gauge spring‐loaded device with 4‐5 core samples obtained during each procedure. All biopsies were well tolerated without significant pain or post‐procedure complications.DiscussionUltrasound‐guided percutaneous core‐needle testicular biopsy using spermatic cord block is a safe and effective option in sampling indeterminate testicular masses while maintaining patient comfort.ConclusionThe inclusion of a spermatic cord block in combination with local anaesthetic and moderate sedation has become standard practice in our institution, as we believe this maximises patient comfort and safety resulting in a better patient experience.

Background/Objective: To share technical pearls of a basic knee arthroscopy in a pediatric patient by presenting tips and tricks used by experienced pediatric arthroscopists. Indications: Knee arthroscopy is a common procedure performed by most orthopedic surgeons. However, technical considerations and equipment should be considered before performing this procedure in a pediatric patient. This video provides information on indications, equipment, and patient positioning, as well as outlining a systematic approach to diagnostic arthroscopy and pertinent pediatric anatomy. Discussion/Conclusion: This video provides an excellent overview as well as specific pathology-dependent considerations for the inexperienced orthopedic surgeon, or the orthopedic surgeon who rarely performs knee arthroscopy. Patient Consent Disclosure Statement: The author(s) attests that consent has been obtained from any patient(s) appearing in this publication. If the individual may be identifiable, the author(s) has included a statement of release or other written form of approval from the patient(s) with this submission for publication.

Importance Electroencephalograms (EEGs) are a fundamental evaluation in neurology but require special expertise unavailable in many regions of the world. Artificial intelligence (AI) has a potential for addressing these unmet needs. Previous AI models address only limited aspects of EEG interpretation such as distinguishing abnormal from normal or identifying epileptiform activity. A comprehensive, fully automated interpretation of routine EEG based on AI suitable for clinical practice is needed. Objective To develop and validate an AI model (Standardized Computer-based Organized Reporting of EEG–Artificial Intelligence [SCORE-AI]) with the ability to distinguish abnormal from normal EEG recordings and to classify abnormal EEG recordings into categories relevant for clinical decision-making: epileptiform-focal, epileptiform-generalized, nonepileptiform-focal, and nonepileptiform-diffuse. Design, Setting, and Participants In this multicenter diagnostic accuracy study, a convolutional neural network model, SCORE-AI, was developed and validated using EEGs recorded between 2014 and 2020. Data were analyzed from January 17, 2022, until November 14, 2022. A total of 30 493 recordings of patients referred for EEG were included into the development data set annotated by 17 experts. Patients aged more than 3 months and not critically ill were eligible. The SCORE-AI was validated using 3 independent test data sets: a multicenter data set of 100 representative EEGs evaluated by 11 experts, a single-center data set of 9785 EEGs evaluated by 14 experts, and for benchmarking with previously published AI models, a data set of 60 EEGs with external reference standard. No patients who met eligibility criteria were excluded. Main Outcomes and Measures Diagnostic accuracy, sensitivity, and specificity compared with the experts and the external reference standard of patients’ habitual clinical episodes obtained during video-EEG recording. Results The characteristics of the EEG data sets include development data set (N = 30 493; 14 980 men; median age, 25.3 years [95% CI, 1.3-76.2 years]), multicenter test data set (N = 100; 61 men, median age, 25.8 years [95% CI, 4.1-85.5 years]), single-center test data set (N = 9785; 5168 men; median age, 35.4 years [95% CI, 0.6-87.4 years]), and test data set with external reference standard (N = 60; 27 men; median age, 36 years [95% CI, 3-75 years]). The SCORE-AI achieved high accuracy, with an area under the receiver operating characteristic curve between 0.89 and 0.96 for the different categories of EEG abnormalities, and performance similar to human experts. Benchmarking against 3 previously published AI models was limited to comparing detection of epileptiform abnormalities. The accuracy of SCORE-AI (88.3%; 95% CI, 79.2%-94.9%) was significantly higher than the 3 previously published models ( P < .001) and similar to human experts. Conclusions and Relevance In this study, SCORE-AI achieved human expert level performance in fully automated interpretation of routine EEGs. Application of SCORE-AI may improve diagnosis and patient care in underserved areas and improve efficiency and consistency in specialized epilepsy centers.

11529 Background: To evaluate clinical characteristics and patterns of treatment failure in patients with localized Ewing sarcoma (ES) treated on AEWS1031 according to local control (LC) strategies, tumor size, and tumor site. Methods: AEWS1031 was a phase 3 randomized trial comparing two interval compressed chemotherapy regimens. Patients who completed LC on AEWS1031 were analyzed. LC was with surgery alone (S), definitive radiation therapy (RT), or surgery plus radiation (S+RT), and determined by the treating investigator. Local failure (LF) and distant failure (DF) were defined as recurrence at primary tumor site or distant site, respectively. Fine and Gray method was used to estimate cumulative incidence of LF and DF from time LC was completed. P-values were calculated using logrank or Gray’s test, as appropriate. A two-sided p-value of ≤0.05 was considered significant. Results: 588 patients completed LC. Median age at enrollment was 13.0 years (range: 0.6- 33.9 years). Tumor sites were categorized as extremity (230; 39%), axial (142; 24%), extraosseous (110; 19%), and pelvis (106; 18%). At diagnosis, tumor volume (TV) was ≥200 mL in 170 patients (31%) and maximum tumor dimension (MTD) was ≥8 cm in 289 patients (52%). LC was with S in 320 patients (54%), RT in 160 (27%), and S+RT in 108 (18%). Eight patients received preoperative RT. Fifty-three patients (13.1%) who received S±RT had an R1 resection. LC with S was more likely for extremity (58%), RT for pelvis (38%) and S+ RT for axial (36%) and extraosseous (34%) primary tumors (p <0.01). With median follow up of 67.6 months from LC, the 5-year cumulative incidences of LF and DF for the entire cohort were 6.0% (95% CI, 4.3-8.3%) and 11.0% (95% CI, 8.6-13.9%), respectively. Eleven patients experienced simultaneous LF and DF. LF incidence was 5.0% for S, 8.4% for RT, and 5.6% for S+RT (p=0.47). DF incidence was 11.8% for S, 6.4% for RT, and 15.1% for S+RT (p=0.16). LF incidence by tumor site was 3.6% for extremity, 8.7% for pelvis, 7.3% for axial, and 6.8% for extraosseous (p=0.08). LF incidence was higher for primary tumors ≥200 mL (11.3%) compared to tumors < 200 mL (3.9%; p<0.01) and for tumors ≥8 cm (7.8%) compared to tumors < 8 cm (4.4%; p=0.02). Tumor size was associated with a higher LF incidence for S and RT, but not for S+RT (Table). LF incidence was 4.0% for R0, 7.5% for R1 with RT, and 16.7% for R1 without RT (p=0.02). Conclusions: We report the lowest LF incidence to date for prospective ES trials conducted. Both larger TV and MTD were associated with higher LF. Future investigations will aim to evaluate the impact of chemotherapy dose-intensity on LC and identify risk-stratification variables for improved LC. Five-year LF incidence by LC and tumor size. Clinical trial information: NCT01231906 . [Table: see text]

Abstract Epilepsy neuroimaging assessment requires exceptional anatomic detail, physiologic and metabolic information. Magnetic resonance (MR) protocols are often time-consuming necessitating sedation and positron emission tomography (PET)/computed tomography (CT) comes with a significant radiation dose. Hybrid PET/MRI protocols allow for exquisite assessment of brain anatomy and structural abnormalities, in addition to metabolic information in a single, convenient imaging session, which limits radiation dose, sedation time, and sedation events. Brain PET/MRI has proven especially useful for accurate localization of epileptogenic zones in pediatric seizure cases, providing critical additional information and guiding surgical decision making in medically refractory cases. Accurate localization of seizure focus is necessary to limit the extent of the surgical resection, preserve healthy brain tissue, and achieve seizure control. This review provides a systematic overview with illustrative examples demonstrating the applications and diagnostic utility of PET/MRI in pediatric epilepsy.