Comparative Accuracy of Diagnosis by Collective Intelligence of Multiple Physicians vs Individual Physicians

Approach to Acute Dizziness/Vertigo in the Emergency Department: Selected Controversies Regarding Specialty Consultation

Acute dizziness and vertigo are common emergency department presentations (≈4% of annual visits) and sometimes, a life-threatening diagnosis like stroke is missed. Recent literature reviews the challenges in evaluation of these symptoms and offers guidelines for diagnostic approaches. Strong evidence indicates that when well-trained providers perform a high-quality bedside neurovestibular examination, accurate diagnosis of peripheral vestibular disorders and stroke increases. However, it is less clear who can and should be performing these assessments on a routine basis. This article offers a focused debate for and against routine specialty consultation for patients with acute dizziness or vertigo in the emergency department as well as a potential path forward utilizing new portable technologies to quantify eye movements.

Root cause analysis of cases involving diagnosis

Diagnostic errors comprise the leading threat to patient safety in healthcare today. Learning how to extract the lessons from cases where diagnosis succeeds or fails is a promising approach to improve diagnostic safety going forward. We present up-to-date and authoritative guidance on how the existing approaches to conducting root cause analyses (RCA's) can be modified to study cases involving diagnosis. There are several diffierences: In cases involving diagnosis, the investigation should begin immediately after the incident, and clinicians involved in the case should be members of the RCA team. The review must include consideration of how the clinical reasoning process went astray (or succeeded), and use a human-factors perspective to consider the system-related contextual factors in the diagnostic process. We present detailed instructions for conducting RCA's of cases involving diagnosis, with advice on how to identify root causes and contributing factors and select appropriate interventions.

Multimodal assessment improves neuroprognosis performance in clinically unresponsive critical-care patients with brain injury

Accurately predicting functional outcomes for unresponsive patients with acute brain injury is a medical, scientific and ethical challenge. This prospective study assesses how a multimodal approach combining various numbers of behavioral, neuroimaging and electrophysiological markers affects the performance of outcome predictions. We analyzed data from 349 patients admitted to a tertiary neurointensive care unit between 2009 and 2021, categorizing prognoses as good, uncertain or poor, and compared these predictions with observed outcomes using the Glasgow Outcome Scale-Extended (GOS-E, levels ranging from 1 to 8, with higher levels indicating better outcomes). After excluding cases with life-sustaining therapy withdrawal to mitigate the self-fulfilling prophecy bias, our findings reveal that a good prognosis, compared with a poor or uncertain one, is associated with better one-year functional outcomes (common odds ratio (95% CI) for higher GOS-E: OR = 14.57 (5.70-40.32), P < 0.001; and 2.9 (1.56-5.45), P < 0.001, respectively). Moreover, increasing the number of assessment modalities decreased uncertainty (OR = 0.35 (0.21-0.59), P < 0.001) and improved prognostic accuracy (OR = 2.72 (1.18-6.47), P = 0.011). Our results underscore the value of multimodal assessment in refining neuroprognostic precision, thereby offering a robust foundation for clinical decision-making processes for acutely brain-injured patients. ClinicalTrials.gov registration: NCT04534777 .

Diagnostic errors and characteristics of patients seen at a general internal medicine outpatient clinic with a referral for diagnosis

OBJECTIVES

Patients referred to general internal medicine (GIM) outpatient clinics may face a higher risk of diagnostic errors than non-referred patients. This difference in risk is assumed to be due to the differences in diseases and clinical presentations between referred and non-referred patients; however, clinical data regarding this issue are scarce. This study aimed to determine the frequency of diagnostic errors and compare the characteristics of referred and non-referred patients visit GIM outpatient clinics.

METHODS

This study included consecutive outpatients who visited the GIM outpatient clinic at a university hospital, with or without referral. Data on age, sex, chief complaints, referral origin, and final diagnosis were collected from medical records. The Revised Safer Dx Instrument was used to detect diagnostic errors.

RESULTS

Data from 534 referred and 599 non-referred patients were analyzed. The diagnostic error rate was higher in the referral group than that in the non-referral group (2.2 % vs. 0.5 %, p=0.01). The prevalence of abnormal test results and sensory disturbances was higher in the chief complaints, and the prevalence of musculoskeletal system disorders, connective tissue diseases, and neoplasms was higher in the final diagnoses of referred patients compared with non-referred patients. Among referred patients with diagnostic errors, abnormal test results and sensory disturbances were the two most common chief complaints, whereas neoplasia was the most common final diagnosis. Problems with data integration and interpretation were found to be the most common factors contributing to diagnostic errors.

CONCLUSIONS

Paying more attention to patients with abnormal test results and sensory disturbances and considering a higher pre-test probability for neoplasms may prevent diagnostic errors in patients referred to GIM outpatient clinics.

Enhancing human-AI collaboration: The case of colonoscopy

Diagnostic errors impact patient health and healthcare costs. Artificial Intelligence (AI) shows promise in mitigating this burden by supporting Medical Doctors in decision-making. However, the mere display of excellent or even superhuman performance by AI in specific tasks does not guarantee a positive impact on medical practice. Effective AI assistance should target the primary causes of human errors and foster effective collaborative decision-making with human experts who remain the ultimate decision-makers. In this narrative review, we apply these principles to the specific scenario of AI assistance during colonoscopy. By unraveling the neurocognitive foundations of the colonoscopy procedure, we identify multiple bottlenecks in perception, attention, and decision-making that contribute to diagnostic errors, shedding light on potential interventions to mitigate them. Furthermore, we explored how existing AI devices fare in clinical practice and whether they achieved an optimal integration with the human decision-maker. We argue that to foster optimal Human-AI collaboration, future research should expand our knowledge of factors influencing AI's impact, establish evidence-based cognitive models, and develop training programs based on them. These efforts will enhance human-AI collaboration, ultimately improving diagnostic accuracy and patient outcomes. The principles illuminated in this review hold more general value, extending their relevance to a wide array of medical procedures and beyond.

Retrieval-Based Diagnostic Decision Support: Mixed Methods Study

BACKGROUND

Diagnostic errors pose significant health risks and contribute to patient mortality. With the growing accessibility of electronic health records, machine learning models offer a promising avenue for enhancing diagnosis quality. Current research has primarily focused on a limited set of diseases with ample training data, neglecting diagnostic scenarios with limited data availability.

OBJECTIVE

This study aims to develop an information retrieval (IR)-based framework that accommodates data sparsity to facilitate broader diagnostic decision support.

METHODS

We introduced an IR-based diagnostic decision support framework called CliniqIR. It uses clinical text records, the Unified Medical Language System Metathesaurus, and 33 million PubMed abstracts to classify a broad spectrum of diagnoses independent of training data availability. CliniqIR is designed to be compatible with any IR framework. Therefore, we implemented it using both dense and sparse retrieval approaches. We compared CliniqIR's performance to that of pretrained clinical transformer models such as Clinical Bidirectional Encoder Representations from Transformers (ClinicalBERT) in supervised and zero-shot settings. Subsequently, we combined the strength of supervised fine-tuned ClinicalBERT and CliniqIR to build an ensemble framework that delivers state-of-the-art diagnostic predictions.

RESULTS

On a complex diagnosis data set (DC3) without any training data, CliniqIR models returned the correct diagnosis within their top 3 predictions. On the Medical Information Mart for Intensive Care III data set, CliniqIR models surpassed ClinicalBERT in predicting diagnoses with <5 training samples by an average difference in mean reciprocal rank of 0.10. In a zero-shot setting where models received no disease-specific training, CliniqIR still outperformed the pretrained transformer models with a greater mean reciprocal rank of at least 0.10. Furthermore, in most conditions, our ensemble framework surpassed the performance of its individual components, demonstrating its enhanced ability to make precise diagnostic predictions.

CONCLUSIONS

Our experiments highlight the importance of IR in leveraging unstructured knowledge resources to identify infrequently encountered diagnoses. In addition, our ensemble framework benefits from combining the complementary strengths of the supervised and retrieval-based models to diagnose a broad spectrum of diseases.

Medical residents' perceptions of group biases in medical decision making: a qualitative study

BACKGROUND

Systematic biases in group decision making (i.e., group biases) may result in suboptimal decisions and potentially harm patients. It is not well known how impaired group decision making in patient care may affect medical training. This study aimed to explore medical residents' experiences and perspectives regarding impaired group decision making and the role of group biases in medical decision making.

METHODS

This study used a qualitative approach with thematic analysis underpinned by a social constructionist epistemology. Semi-structured interviews of medical residents were conducted at a single internal medicine residency program. Residents were initially asked about their experiences with suboptimal medical decision making as a group or team. Then, questions were targeted to several group biases (groupthink, social loafing, escalation of commitment). Interviews were transcribed and transferred to a qualitative data analysis software. Thematic analysis was conducted to generate major themes within the dataset.

RESULTS

Sixteen interviews with residents revealed five major themes: (1) hierarchical influence on group decision making; (2) group decision making under pressure; (3) post-call challenges in decision making; (4) interactions between teamwork and decision making; and (5) personal and cultural influences in group decision making. Subthemes were also identified for each major theme. Most residents were able to recognize groupthink in their past experiences working with medical teams. Residents perceived social loafing or escalation of commitment as less relevant for medical team decision making.

CONCLUSIONS

Our findings provide unique insights into the complexities of group decision making processes in teaching hospitals. Team hierarchy significantly influenced residents' experiences with group decision making-most group decisions were attributed to consultants or senior team members, while lower ranking team members contributed less and perceived fewer opportunities to engage in group decisions. Other factors such as time constraints on decision making, perceived pressures from other staff members, and challenges associated with post-call days were identified as important barriers to optimal group decision making in patient care. Future studies may build upon these findings to enhance our understanding of medical team decision making and develop strategies to improve group decisions, ultimately leading to higher quality patient care and training.

Around the World in 60 Minutes: How a Virtual Morning Report has Created an International Community for Clinical Reasoning and Medical Education

Problem: Traditionally, clinical reasoning is developed with purposeful exposure to clinical problems through case-based learning and clinical reasoning conferences that harvest a collaborative exchange of information in real-life settings. While virtual platforms have greatly expanded access to remote clinical learning, case-based clinical reasoning opportunities are scarce in low and middle income countries. Intervention: The Clinical Problem Solvers (CPSolvers), a nonprofit organization focused on clinical reasoning education, launched Virtual Morning Report (VMR) during the COVID-19 pandemic. VMR is an open-access, case-based clinical reasoning virtual conference on the Zoom platform modeled after an academic morning report format available to participants worldwide. The authors conducted 17 semi-structured interviews with CPSolvers' VMR participants from 10 different countries to explore the experiences of the international participants of VMR. Context: The CPSolvers was founded by US physicians and has now expanded to include international members throughout all levels of the organization. VMR is open-access to all learners. Preliminary survey data collected from VMR sessions revealed 35% of the attendees were from non-English speaking countries and 53% from non US countries. Impact: Analysis generated four themes that captured the experiences of international participants of VMR: 1) Improving clinical reasoning skills where participants had little to no access to this education or content; 2) Creating a global community from a diverse, safe, and welcoming environment made possible by the virtual platform; 3) Allowing learners to become agents of change by providing tools and skills that are directly applicable in the setting in which they practice medicine; 4) Establishing a global platform, with low barriers to entry and open-access to expertise and quality teaching and content. Study participants agreed with the themes, supporting trustworthiness. Lessons Learned: Findings suggest VMR functions as and has grown into a global community of practice for clinical reasoning. The authors propose strategies and guiding principles based on the identified themes for educators to consider when building effective global learning communities. In an interdependent world where the virtual space eliminates the physical boundaries that silo educational opportunities, emphasis on thoughtful implementation of learning communities in a global context has the potential to reduce medical education disparities in the clinical reasoning space and beyond.

Boosting wisdom of the crowd for medical image annotation using training performance and task features

A crucial bottleneck in medical artificial intelligence (AI) is high-quality labeled medical datasets. In this paper, we test a large variety of wisdom of the crowd algorithms to label medical images that were initially classified by individuals recruited through an app-based platform. Individuals classified skin lesions from the International Skin Lesion Challenge 2018 into 7 different categories. There was a large dispersion in the geographical location, experience, training, and performance of the recruited individuals. We tested several wisdom of the crowd algorithms of varying complexity from a simple unweighted average to more complex Bayesian models that account for individual patterns of errors. Using a switchboard analysis, we observe that the best-performing algorithms rely on selecting top performers, weighting decisions by training accuracy, and take into account the task environment. These algorithms far exceed expert performance. We conclude by discussing the implications of these approaches for the development of medical AI.

Is the Comparator in Your Diagnostic Cost-Effectiveness Model "Standard of Care"? Recommendations from Literature Reviews and Expert Interviews on How to Identify and Operationalize It

OBJECTIVES

This research aimed to develop best-practice recommendations for identifying the "standard of care" (SoC) and integrate it when it is the comparator in diagnostic economic models (SoC comparator).

METHODS

A multi-methods approach comprising 2 pragmatic literature reviews and 9 expert interviews was used. Experts rated their agreement with draft recommendations based on the authors' analysis of the reviews. These were refined iteratively to produce final recommendations.

RESULTS

Fourteen best-practice recommendations are provided. Care pathway mapping (using quantitative, qualitative, or mixed-methods approaches) should be used for identifying the SoC comparator. Guidelines analysis can be integrated with expert opinion to identify pathway variability and discrepancies from clinical practice. For integrating the SoC comparator into the model, recommendations around structure, input sourcing, data aggregation and reporting, input uncertainty, and model variability are presented. For example, modelers should consider that the reference standard is not synonymous with the SoC, and the SoC may not be the only comparator. The comparator limitations should be discussed with clinical experts, but elicitation of its diagnostic accuracy is not recommended. Probabilistic sensitivity analysis is recommended when evaluating the overall input uncertainty, and deterministic sensitivity analysis is useful when there is high model uncertainty or SoC variability. Consensus could not be reached for some topics (eg, the role of real-world data, model averaging, and alternative model structures), but the reported discussions provide points for consideration.

CONCLUSIONS

To our knowledge, this is the first guidance to support modelers when identifying and operationalizing the SoC comparator in diagnostic cost-effectiveness models.

Collective Intelligence Increases Diagnostic Accuracy in a General Practice Setting

BACKGROUND

General practitioners (GPs) work in an ill-defined environment where diagnostic errors are prevalent. Previous research indicates that aggregating independent diagnoses can improve diagnostic accuracy in a range of settings. We examined whether aggregating independent diagnoses can also improve diagnostic accuracy for GP decision making. In addition, we investigated the potential benefit of such an approach in combination with a decision support system (DSS).

METHODS

We simulated virtual groups using data sets from 2 previously published studies. In study 1, 260 GPs independently diagnosed 9 patient cases in a vignette-based study. In study 2, 30 GPs independently diagnosed 12 patient actors in a patient-facing study. In both data sets, GPs provided diagnoses in a control condition and/or DSS condition(s). Each GP's diagnosis, confidence rating, and years of experience were entered into a computer simulation. Virtual groups of varying sizes (range: 3-9) were created, and different collective intelligence rules (plurality, confidence, and seniority) were applied to determine each group's final diagnosis. Diagnostic accuracy was used as the performance measure.

RESULTS

Aggregating independent diagnoses by weighing them equally (i.e., the plurality rule) substantially outperformed average individual accuracy, and this effect increased with increasing group size. Selecting diagnoses based on confidence only led to marginal improvements, while selecting based on seniority reduced accuracy. Combining the plurality rule with a DSS further boosted performance.

DISCUSSION

Combining independent diagnoses may substantially improve a GP's diagnostic accuracy and subsequent patient outcomes. This approach did, however, not improve accuracy in all patient cases. Therefore, future work should focus on uncovering the conditions under which collective intelligence is most beneficial in general practice.

HIGHLIGHTS

We examined whether aggregating independent diagnoses of GPs can improve diagnostic accuracy.Using data sets of 2 previously published studies, we composed virtual groups of GPs and combined their independent diagnoses using 3 collective intelligence rules (plurality, confidence, and seniority).Aggregating independent diagnoses by weighing them equally substantially outperformed average individual GP accuracy, and this effect increased with increasing group size.Combining independent diagnoses may substantially improve GP's diagnostic accuracy and subsequent patient outcomes.

Crowdsourcing a diagnosis? Exploring the accuracy of the size and type of group diagnosis: an experimental study

BACKGROUND

The consultation process, where a clinician seeks an opinion from another clinician, is foundational in medicine. However, the effectiveness of group diagnosis has not been studied.

OBJECTIVE

To compare individual diagnosis to group diagnosis on two dimensions: group size (n=3 or 6) and group process (interactive or artificial groups).

METHODOLOGY

Thirty-six internal or emergency medicine residents participated in the study. Initially, each resident worked through four written cases on their own, providing a primary diagnosis and a differential diagnosis. Next, participants formed into groups of three. Using a videoconferencing platform, they worked through four additional cases, collectively providing a single primary diagnosis and differential diagnosis. The process was repeated using a group of six with four new cases. Cases were all counterbalanced. Retrospectively, nominal (ie, artificial) groups were formed by aggregating individual participant data into subgroups of three and six and analytically computing scores. Presence of the correct diagnosis as primary diagnosis or included in the differential diagnosis, as well as the number of diagnoses mentioned, was calculated for all conditions. Means were compared using analysis of variance.

RESULTS

For both authentic and nominal groups, the diagnostic accuracy of group diagnosis was superior to individual for both the primary diagnosis and differential diagnosis. However, there was no improvement in diagnostic accuracy when comparing a group of three to a group of six. Interactive and nominal groups were equivalent; however, this may be an artefact of the method used to combine data.

CONCLUSIONS

Group diagnosis improves diagnostic accuracy. However, a larger group is not necessarily superior to a smaller group. In this study, interactive group discussion does not result in improved diagnostic accuracy.

Sustainable reduction of phone-call interruptions by 35% in a medical imaging department using an automatic voicemail and custom call redirection system

BACKGROUND

Have you ever been in the trenches of a complicated study only to be interrupted by a not-so urgent phone-call? We were, repeatedly- unfortunately.

PURPOSE

To increase productivity of radiologists by quantifying the main source of interruptions (phone-calls) to the workflow of radiologists, and too assess the implemented solution.

MATERIALS AND METHODS

To filter calls to the radiology consultant on duty, we introduced an automatic voicemail and custom call redirection system. Thus, instead of directly speaking with radiology consultants, clinicians were to first categorize their request and dial accordingly: 1. Inpatient requests, 2. Outpatient requests, 3. Directly speak with the consultant radiologist. Inpatient requests (1) and outpatient requests (2) were forwarded to MRI technologists or clerks, respectively. Calls were monitored in 15-minute increments continuously for an entire year (March 2022 until and including March 2023). Subsequently, both the frequency and category of requests were assessed.

RESULTS

4803 calls were recorded in total: 3122 (65 %) were forwarded to a radiologist on duty. 870 (18.11 %) concerned inpatients, 274 (5.70 %) outpatients, 430 (8.95 %) dialed the wrong number, 107 (2.23 %) made no decision. Throughout the entire year the percentage of successfully avoided interruptions was relatively stable and fluctuated between low to high 30 % range (Mean per month 35 %, Median per month 34.45 %).

CONCLUSIONS

This is the first analysis of phone-call interruptions to consultant radiologists in an imaging department for 12 continuous months. More than 35 % of requests did not require the input of a specialist trained radiologist. Hence, installing an automated voicemail and custom call redirection system is a sustainable and simple solution to reduce phone-call interruptions by on average 35 % in radiology departments. This solution was well accepted by referring clinicians. The installation required a one-time investment of only 2h and did not cost any money.

Attaining Diagnostic Excellence: How the Structure and Function of a Rare Disease Service Contribute to Ending the Diagnostic Odyssey

Patients with rare or otherwise undiagnosed disorders frequently find themselves on a diagnostic odyssey, the often-prolonged journey toward diagnosis that can be characterized by significant physical, emotional, and financial hardship, as well as by diagnostic errors and delays. The wider availability of clinical exome sequencing has helped end many diagnostic odysseys, though diagnostic success rates of around 35% for exome sequencing leave many patients undiagnosed. Diagnostic yields can be improved via the implementation of advanced genetic testing modalities, though both these modalities and exome sequencing perform significantly better when paired with high-quality phenotypic data. Diagnostic centers of excellence can improve outcomes for patients on a diagnostic odyssey by providing a process and environment that address shortfalls in diagnostic access while providing high-quality phenotyping. Features of successful undiagnosed and rare disease evaluation teams are discussed and an illustrative case is provided.

PREDICT-TBI: Comparison of Physician Predictions with the IMPACT Model to Predict 6-Month Functional Outcome in Traumatic Brain Injury

BACKGROUND

Predicting functional outcome in critically ill patients with traumatic brain injury (TBI) strongly influences end-of-life decisions and information for surrogate decision makers. Despite well-validated prognostic models, clinicians most often rely on their subjective perception of prognosis. In this study, we aimed to compare physicians' predictions with the International Mission on Prognosis and Analysis of Clinical Trials in TBI (IMPACT) prognostic model for predicting an unfavorable functional outcome at 6 months after moderate or severe TBI.

METHODS

PREDICT-TBI is a prospective study of patients with moderate to severe TBI. Patients were admitted to a neurocritical care unit and were excluded if they died or had withdrawal of life-sustaining treatments within the first 24 h. In a paired study design, we compared the accuracy of physician prediction on day 1 with the prediction of the IMPACT model as two diagnostic tests in predicting unfavorable outcome 6 months after TBI. Unfavorable outcome was assessed by the Glasgow Outcome Scale from 1 to 3 by using a structured telephone interview. The primary end point was the difference between the discrimination ability of the physician and the IMPACT model assessed by the area under the curve.

RESULTS

Of the 93 patients with inclusion and exclusion criteria, 80 patients reached the primary end point. At 6 months, 29 patients (36%) had unfavorable outcome. A total of 31 clinicians participated in the study. Physicians' predictions showed an area under the curve of 0.79 (95% confidence interval 0.68-0.89), against 0.80 (95% confidence interval 0.69-0.91) for the laboratory IMPACT model, with no statistical difference (p = 0.88). Both approaches were well calibrated. Agreement between physicians was moderate (κ = 0.56). Lack of experience was not associated with prediction accuracy (p = 0.58).

CONCLUSIONS

Predictions made by physicians for functional outcome were overall moderately accurate, and no statistical difference was found with the IMPACT models, possibly due to a lack of power. The significant variability between physician assessments suggests prediction could be improved through peer reviewing, with the support of the IMPACT models, to provide a realistic expectation of outcome to families and guide discussions about end-of-life decisions.

Automating hybrid collective intelligence in open-ended medical diagnostics

Collective intelligence has emerged as a powerful mechanism to boost decision accuracy across many domains, such as geopolitical forecasting, investment, and medical diagnostics. However, collective intelligence has been mostly applied to relatively simple decision tasks (e.g., binary classifications). Applications in more open-ended tasks with a much larger problem space, such as emergency management or general medical diagnostics, are largely lacking, due to the challenge of integrating unstandardized inputs from different crowd members. Here, we present a fully automated approach for harnessing collective intelligence in the domain of general medical diagnostics. Our approach leverages semantic knowledge graphs, natural language processing, and the SNOMED CT medical ontology to overcome a major hurdle to collective intelligence in open-ended medical diagnostics, namely to identify the intended diagnosis from unstructured text. We tested our method on 1,333 medical cases diagnosed on a medical crowdsourcing platform: The Human Diagnosis Project. Each case was independently rated by ten diagnosticians. Comparing the diagnostic accuracy of single diagnosticians with the collective diagnosis of differently sized groups, we find that our method substantially increases diagnostic accuracy: While single diagnosticians achieved 46% accuracy, pooling the decisions of ten diagnosticians increased this to 76%. Improvements occurred across medical specialties, chief complaints, and diagnosticians' tenure levels. Our results show the life-saving potential of tapping into the collective intelligence of the global medical community to reduce diagnostic errors and increase patient safety.

Effects of Combinational Use of Additional Differential Diagnostic Generators on the Diagnostic Accuracy of the Differential Diagnosis List Developed by an Artificial Intelligence-Driven Automated History-Taking System: Pilot Cross-Sectional Study

BACKGROUND

Low diagnostic accuracy is a major concern in automated medical history-taking systems with differential diagnosis (DDx) generators. Extending the concept of collective intelligence to the field of DDx generators such that the accuracy of judgment becomes higher when accepting an integrated diagnosis list from multiple people than when accepting a diagnosis list from a single person may be a possible solution.

OBJECTIVE

The purpose of this study is to assess whether the combined use of several DDx generators improves the diagnostic accuracy of DDx lists.

METHODS

We used medical history data and the top 10 DDx lists (index DDx lists) generated by an artificial intelligence (AI)-driven automated medical history-taking system from 103 patients with confirmed diagnoses. Two research physicians independently created the other top 10 DDx lists (second and third DDx lists) per case by imputing key information into the other 2 DDx generators based on the medical history generated by the automated medical history-taking system without reading the index lists generated by the automated medical history-taking system. We used the McNemar test to assess the improvement in diagnostic accuracy from the index DDx lists to the three types of combined DDx lists: (1) simply combining DDx lists from the index, second, and third lists; (2) creating a new top 10 DDx list using a 1/n weighting rule; and (3) creating new lists with only shared diagnoses among DDx lists from the index, second, and third lists. We treated the data generated by 2 research physicians from the same patient as independent cases. Therefore, the number of cases included in analyses in the case using 2 additional lists was 206 (103 cases × 2 physicians' input).

RESULTS

The diagnostic accuracy of the index lists was 46% (47/103). Diagnostic accuracy was improved by simply combining the other 2 DDx lists (133/206, 65%, P<.001), whereas the other 2 combined DDx lists did not improve the diagnostic accuracy of the DDx lists (106/206, 52%, P=.05 in the collective list with the 1/n weighting rule and 29/206, 14%, P<.001 in the only shared diagnoses among the 3 DDx lists).

CONCLUSIONS

Simply adding each of the top 10 DDx lists from additional DDx generators increased the diagnostic accuracy of the DDx list by approximately 20%, suggesting that the combinational use of DDx generators early in the diagnostic process is beneficial.

Experimental evidence for structured information-sharing networks reducing medical errors

Errors in clinical decision-making are disturbingly common. Recent studies have found that 10 to 15% of all clinical decisions regarding diagnoses and treatment are inaccurate. Here, we experimentally study the ability of structured information-sharing networks among clinicians to improve clinicians' diagnostic accuracy and treatment decisions. We use a pool of 2,941 practicing clinicians recruited from around the United States to conduct 84 independent group-level trials, ranging across seven different clinical vignettes for topics known to exhibit high rates of diagnostic or treatment error (e.g., acute cardiac events, geriatric care, low back pain, and diabetes-related cardiovascular illness prevention). We compare collective performance in structured information-sharing networks to collective performance in independent control groups, and find that networks significantly reduce clinical errors, and improve treatment recommendations, as compared to control groups of independent clinicians engaged in isolated reflection. Our results show that these improvements are not a result of simple regression to the group mean. Instead, we find that within structured information-sharing networks, the worst clinicians improved significantly while the best clinicians did not decrease in quality. These findings offer implications for the use of social network technologies to reduce errors among clinicians.

Harnessing the power of collective intelligence in dentistry: a pilot study in Victoria, Australia

BACKGROUND

In many dental settings, diagnosis and treatment planning is the responsibility of a single clinician, and this process is inevitably influenced by the clinician's own heuristics and biases. Our aim was to test whether collective intelligence increases the accuracy of individual diagnoses and treatment plans, and whether such systems have potential to improve patient outcomes in a dental setting.

METHODS

This pilot project was carried out to assess the feasibility of the protocol and appropriateness of the study design. We used a questionnaire survey and pre-post study design in which dental practitioners were involved in the diagnosis and treatment planning of two simulated cases. Participants were provided the opportunity to amend their original diagnosis/treatment decisions after viewing a consensus report made to simulate a collaborative setting.

RESULTS

Around half (55%, n = 17) of the respondents worked in group private practices, however most practitioners (74%, n = 23) did not collaborate when planning treatment. Overall, the average practitioners' self-confidence score in managing different dental disciplines was 7.22 (s.d. 2.20) on a 1-10 scale. Practitioners tended to change their mind after viewing the consensus response, particularly for the complex case compared to the simple case (61.5% vs 38.5%, respectively). Practitioners' confidence ratings were also significantly higher (p < 0.05) after viewing the consensus for complex case.

CONCLUSION

Our pilot study shows that collective intelligence in the form of peers' opinion can lead to modifications in diagnosis and treatment planning by dentists. Our results lay the foundations for larger scale investigations on whether peer collaboration can improve diagnostic accuracy, treatment planning and, ultimately, oral health outcomes.

Heuristics and biases in medical decision-making under uncertainty: The case of neuropronostication for consciousness disorders

Neuropronostication for consciousness disorders can be very complex and prone to high uncertainty. Despite notable advancements in the development of dedicated scales and physiological markers using innovative paradigms, these technical progressions are often overshadowed by factors intrinsic to the medical environment. Beyond the scarcity of objective data guiding medical decisions, factors like time pressure, fatigue, multitasking, and emotional load can drive clinicians to rely more on heuristic-based clinical reasoning. Such an approach, albeit beneficial under certain circumstances, may lead to systematic error judgments and impair medical decisions, especially in complex and uncertain environments. After a brief review of the main theoretical frameworks, this paper explores the influence of clinicians' cognitive biases on clinical reasoning and decision-making in the challenging context of neuroprognostication for consciousness disorders. The discussion further revolves around developing and implementing various strategies designed to mitigate these biases and their impact, aiming to enhance the quality of care and the patient safety.

Collective intelligence improves probabilistic diagnostic assessments

OBJECTIVES

Collective intelligence, the "wisdom of the crowd," seeks to improve the quality of judgments by aggregating multiple individual inputs. Here, we evaluate the success of collective intelligence strategies applied to probabilistic diagnostic judgments.

METHODS

We compared the performance of individual and collective intelligence judgments on two series of clinical cases requiring probabilistic diagnostic assessments, or "forecasts". We assessed the quality of forecasts using Brier scores, which compare forecasts to observed outcomes.

RESULTS

On both sets of cases, the collective intelligence answers outperformed nearly every individual forecaster or team. The improved performance by collective intelligence was mediated by both improved resolution and calibration of probabilistic assessments. In a secondary analysis looking at the effect of varying number of individual inputs in collective intelligence answers from two different data sources, nearly identical curves were found in the two data sets showing 11-12% improvement when averaging two independent inputs, 15% improvement averaging four independent inputs, and small incremental improvements with further increases in number of individual inputs.

CONCLUSIONS

Our results suggest that the application of collective intelligence strategies to probabilistic diagnostic forecasts is a promising approach to improve diagnostic accuracy and reduce diagnostic error.

Diagnostic Accuracy of Differential-Diagnosis Lists Generated by Generative Pretrained Transformer 3 Chatbot for Clinical Vignettes with Common Chief Complaints: A Pilot Study

The diagnostic accuracy of differential diagnoses generated by artificial intelligence (AI) chatbots, including the generative pretrained transformer 3 (GPT-3) chatbot (ChatGPT-3) is unknown. This study evaluated the accuracy of differential-diagnosis lists generated by ChatGPT-3 for clinical vignettes with common chief complaints. General internal medicine physicians created clinical cases, correct diagnoses, and five differential diagnoses for ten common chief complaints. The rate of correct diagnosis by ChatGPT-3 within the ten differential-diagnosis lists was 28/30 (93.3%). The rate of correct diagnosis by physicians was still superior to that by ChatGPT-3 within the five differential-diagnosis lists (98.3% vs. 83.3%, p = 0.03). The rate of correct diagnosis by physicians was also superior to that by ChatGPT-3 in the top diagnosis (53.3% vs. 93.3%, p < 0.001). The rate of consistent differential diagnoses among physicians within the ten differential-diagnosis lists generated by ChatGPT-3 was 62/88 (70.5%). In summary, this study demonstrates the high diagnostic accuracy of differential-diagnosis lists generated by ChatGPT-3 for clinical cases with common chief complaints. This suggests that AI chatbots such as ChatGPT-3 can generate a well-differentiated diagnosis list for common chief complaints. However, the order of these lists can be improved in the future.

Predicting pneumonia during hospitalization in flail chest patients using machine learning approaches

Objective

Pneumonia is a common pulmonary complication of flail chest, causing high morbidity and mortality rates in affected patients. The existing methods for identifying pneumonia have low accuracy, and their use may delay antimicrobial therapy. However, machine learning can be combined with electronic medical record systems to identify information and assist in quick clinical decision-making. Our study aimed to develop a novel machine-learning model to predict pneumonia risk in flail chest patients.

Methods

From January 2011 to December 2021, the electronic medical records of 169 adult patients with flail chest at a tertiary teaching hospital in an urban level I Trauma Centre in Chongqing were retrospectively analysed. Then, the patients were randomly divided into training and test sets at a ratio of 7:3. Using the Fisher score, the best subset of variables was chosen. The performance of the seven models was evaluated by computing the area under the receiver operating characteristic curve (AUC). The output of the XGBoost model was shown using the Shapley Additive exPlanation (SHAP) method.

Results

Of 802 multiple rib fracture patients, 169 flail chest patients were eventually included, and 86 (50.80%) were diagnosed with pneumonia. The XGBoost model performed the best among all seven machine-learning models. The AUC of the XGBoost model was 0.895 (sensitivity: 84.3%; specificity: 80.0%).Pneumonia in flail chest patients was associated with several features: systolic blood pressure, pH value, blood transfusion, and ISS.

Conclusion

Our study demonstrated that the XGBoost model with 32 variables had high reliability in assessing risk indicators of pneumonia in flail chest patients. The SHAP method can identify vital pneumonia risk factors, making the XGBoost model's output clinically meaningful.

The Effectiveness of the Multiple-Attending-Physicians System Compared With the Single Attending-Physician System in Inpatient Setting: A Mixed-Method Study

OBJECTIVES

Medical facilities have been required to effectively utilize insufficient human resources in many countries. Therefore, we qualitatively and quantitively compared physicians' working burden, and assessed advantages and disadvantages of the single- and the multiple-attending physicians systems in inpatient care.

METHODS

In this cross-sectional study, we extracted electronic health record of patients from a hospital in Japan from April 2017 to October 2018 to compare anonymous statistical data between the single-attending and multiple-attending-physicians system. Then, we conducted a questionnaire survey for all physicians of single and multiple-attending systems, asking about their physical and psychiatric workload, and their reasons and comments on their working styles.

RESULTS

The average length of hospital stay was significantly shorter in the multiple-attending system than in the single-attending system, while patients' age, gender, and diagnoses were similar. From the questionnaire survey, no significant difference was found in all categories although physical burden in multiple-attending system tended to be lower than that in single-attending system. Advantages of multiple-attending system extracted from qualitative analysis are (1) improvement of physicians' quality of life (QOL), (2) lifelong-learning effect, and (3) improving the quality of medical care, while disadvantages were (1) risk of miscommunications, (2) conflicting treatment policies among physicians, and (3) patients' concern.

CONCLUSIONS

The multiple-attending physician system in the inpatient setting can reduce the average length of stay for patients and also reduce the physical burden on physicians without compromising their clinical performance.

Issues in Melanoma Detection: Semisupervised Deep Learning Algorithm Development via a Combination of Human and Artificial Intelligence

BACKGROUND

Automatic skin lesion recognition has shown to be effective in increasing access to reliable dermatology evaluation; however, most existing algorithms rely solely on images. Many diagnostic rules, including the 3-point checklist, are not considered by artificial intelligence algorithms, which comprise human knowledge and reflect the diagnosis process of human experts.

OBJECTIVE

In this paper, we aimed to develop a semisupervised model that can not only integrate the dermoscopic features and scoring rule from the 3-point checklist but also automate the feature-annotation process.

METHODS

We first trained the semisupervised model on a small, annotated data set with disease and dermoscopic feature labels and tried to improve the classification accuracy by integrating the 3-point checklist using ranking loss function. We then used a large, unlabeled data set with only disease label to learn from the trained algorithm to automatically classify skin lesions and features.

RESULTS

After adding the 3-point checklist to our model, its performance for melanoma classification improved from a mean of 0.8867 (SD 0.0191) to 0.8943 (SD 0.0115) under 5-fold cross-validation. The trained semisupervised model can automatically detect 3 dermoscopic features from the 3-point checklist, with best performances of 0.80 (area under the curve [AUC] 0.8380), 0.89 (AUC 0.9036), and 0.76 (AUC 0.8444), in some cases outperforming human annotators.

CONCLUSIONS

Our proposed semisupervised learning framework can help with the automatic diagnosis of skin disease based on its ability to detect dermoscopic features and automate the label-annotation process. The framework can also help combine semantic knowledge with a computer algorithm to arrive at a more accurate and more interpretable diagnostic result, which can be applied to broader use cases.

[Healthcare networks for people with rare diseases: integrating data and expertise]

In the European Union (EU), rare diseases (RDs) are diseases that affect no more than 5 in 10,000 people. Due to their rarity, clinical expertise and quality-assured care structures are scarce, and research is more difficult compared to other diseases. However, these problems can be overcome by means of national and transnational RD care networks. Data and expertise are pooled in these networks.In the EU, the European Reference Networks (ERNs) for Rare and Complex Diseases cooperate across borders. Important services provided by ERNs using health data include diagnostic coding of RDs, conducting virtual cross-border case conferences, and establishing European registries that are used to measure and improve the quality of care. In ERNs, local data generation and documentation combine with network-wide data infrastructures. This paper describes the data-based services in and for RD healthcare networks: (1) diagnostic coding, (2) cross-border case conferences, and (3) ERN registries for RD patient care. The final section discusses the integration of the networks into national healthcare systems.In order to achieve the best possible benefit for SE patients, ERN activities and structures need to be better integrated into national healthcare systems. In Germany, the Medical Informatics Initiative and the German Reference Networks play a central role in this regard.

The network science of collective intelligence

In the last few years, breakthroughs in computational and experimental techniques have produced several key discoveries in the science of networks and human collective intelligence. This review presents the latest scientific findings from two key fields of research: collective problem-solving and the wisdom of the crowd. I demonstrate the core theoretical tensions separating these research traditions and show how recent findings offer a new synthesis for understanding how network dynamics alter collective intelligence, both positively and negatively. I conclude by highlighting current theoretical problems at the forefront of research on networked collective intelligence, as well as vital public policy challenges that require new research efforts.

Assigning a different endoscopist for each annual follow-up may contribute to improved gastric cancer detection rates

Background and study aims  Esophagogastroduodenoscopy (EGD) is an effective and important diagnostic tool to detect gastric cancer (GC). Although previous studies show that examiner, patient, and instrumental factors influence the detection of GC, we analyzed whether assigning a different examiner to surveillance EGD would improve the detection of GC compared to assigning the same examiner as in the previous endoscopy. Patients and methods  We retrospectively reviewed patients who underwent two or more consecutive surveillance EGDs at a single center between 2017 and 2019. We identified factors associated with GC detection using multivariable regression analysis and propensity-score matching. Results  Among 7794 patients, 99 GC lesions in 93 patients were detected by surveillance EGD (detection rate; 1.2 %), with a mean surveillance interval of 11.2 months. Among the detected 99 lesions, 87 (87.9 %) were curatively treated with endoscopy. There were no differences in the clinicopathologic characteristics of GC detected by the same or different endoscopists. GC detection in the group examined by different endoscopists was more statistically significant than in the group examined by the same endoscopist, even after propensity-score matching (1.6 % and 0.7 %; P  < 0.05). Endoscopic experience and other factors were not statistically significant between the two groups. Conclusions  In surveillance EGD, having a different endoscopist for each exam may improve GC detection rates, regardless of the endoscopist's experience.

Psychiatric diagnostic dilemmas among people with intellectual and developmental disabilities

BACKGROUND

Research regarding the accuracy of co-morbid psychiatric diagnoses in individuals with intellectual and developmental disabilities (IDD) is sparse. Yet correct diagnostic assignment is vital so that effective and appropriate treatment can be implemented, especially for the large numbers of individuals requiring expensive and restrictive behavioural health crisis services.

METHOD

A retrospective review of de-identified data from multidisciplinary specialty team assessments completed for 50 individuals with ID (IntellectualDisability) with and without ASD and unresolved behavioural health challenges was conducted. The accuracy and reliability of the psychiatric diagnoses upon referral were compared with the diagnoses after the comprehensive team evaluation, and within-individual diagnostic agreement was calculated. The agreement between the Mood and Anxiety Semi-Structured interview tool (MASS) and the full team evaluation was also calculated. The influence of demographic and clinical characteristics on diagnostic agreement was explored.

RESULTS

The most common chief complaints upon referral were aggression to others and self-injurious behaviour. Individuals were taking a median of six medications (interquartile range: 5 to 7); 80% were taking an antipsychotic medication. The most common medical conditions were constipation (70%) and gastroesophageal reflux disease (52%). Measures of interrater reliability of the referral diagnoses with the team assessment were below 0.5 (kappa range: -0.04 to 0.39), with the exception of ruling out dementia (kappa = 0.85). The interrater reliability estimates for the MASS evaluations for depression and anxiety were higher (kappa = 0.69 and 0.64) and reflected higher sensitivity and PPV. The odds of any referral diagnosis being confirmed by team evaluation were low: 0.25 (range: 0 to 0.67). The level of diagnostic agreement for each patient was not significantly attributable to demographic or clinical characteristics, although effect sizes indicate a possible positive relationship to age and the number of prescribed psychotropic medications at referral.

CONCLUSION

Individuals in the current study had serious psychiatric and behavioural problems despite psychiatric care in their communities. The majority of psychiatric diagnoses provided upon referral were not supported by the multidisciplinary specialty team's assessment. In addition to possible diagnostic inaccuracy, the group in the study suffered from multiple medical co-morbidities and were exposed to polypharmacy. Results emphasise the importance of multidisciplinary evaluation by clinicians with expertise in neurodevelopmental disabilities when people with ID with and without ASD have complex behavioural health needs that are unresponsive to usual care. In addition, based on agreement with the full team evaluation, the MASS shows promise as an assessment tool, especially with regards to identifying anxiety and depression.

Clinicians' Use of Quantitative Information When Assessing the Rate of Structural Progression in Glaucoma

PURPOSE

OCT scans contain large amounts of information, but clinicians often rely on reported layer thicknesses when assessing the rate of glaucomatous progression. We sought to determine which of these quantifications most closely relate to the subjective assessment of glaucoma experts who had all the diagnostic information available.

DESIGN

Prospective cohort study.

PARTICIPANTS

Eleven glaucoma specialists independently scored the rate of structural progression from a series of 5 biannual clinical OCT printouts.

METHODS

A total of 100 glaucoma or glaucoma suspect eyes of 51 participants were included; 20 were scored twice to assess repeatability. Scores ranged from 1 (improvement) to 7 (very rapid progression). Generalized estimating equation linear models were used to predict the mean clinician score from the rates of change of retinal nerve fiber layer thickness (RNFLT) or minimum rim width (MRW) globally or in the most rapidly thinning of the 6 sectors.

MAIN OUTCOME MEASURES

The correlation between the objective rates of change and the average of the 11 clinicians' scores.

RESULTS

Average RNFLT within the series of study eyes was 79.3 μm (range, 41.4-126.6). Some 95% of individual clinician scores varied by ≤ 1 point when repeated. The mean clinician score was more strongly correlated with the rate of change of RNFLT in the most rapidly changing sector in %/year (pseudo-R2 = 0.657) than the rate of global RNFLT (0.372). The rate of MRW in the most rapidly changing sector had pseudo-R2 = 0.149.

CONCLUSIONS

The rate of change of RNFLT in the most rapidly changing sector predicted experts' assessment of the rate of structural progression better than global rates or MRW. Sectoral rates may be a useful addition to current clinical printouts.

The Cases for and against Artificial Intelligence in the Medical School Curriculum

Although artificial intelligence (AI) has immense potential to shape the future of medicine, its place in undergraduate medical education currently is unclear. Numerous arguments exist both for and against including AI in the medical school curriculum. AI likely will affect all medical specialties, perhaps radiology more so than any other. The purpose of this article is to present a balanced perspective on whether AI should be included officially in the medical school curriculum. After presenting the balanced point-counterpoint arguments, the authors provide a compromise. Keywords: Artificial Intelligence, Medical Education, Medical School Curriculum, Medical Students, Radiology, Use of AI in Education © RSNA, 2022.

DenseNet-II: an improved deep convolutional neural network for melanoma cancer detection

Research in the field of medicine and relevant studies evince that melanoma is one of the deadliest cancers. It defines precisely that the condition develops due to uncontrolled growth of melanocytic cells. The current trends in any disease detection revolve around the usage of two main categories of models; these are general machine learning models and deep learning models. Further, the experimental analysis of melanoma has an additional requirement of visual records like dermatological scans or normal camera lens images. This further accentuates the need for a more accurate model for melanoma detection. In this work, we aim to achieve the same, primarily by the extensive usage of neural networks. Our objective is to propose a deep learning CNN framework-based model to improve the accuracy of melanoma detection by customizing the number of layers in the network architecture, activation functions applied, and the dimension of the input array. Models like Resnet, DenseNet, Inception, and VGG have proved to yield appreciable accuracy in melanoma detection. However, in most cases, the dataset was classified into malignant or benign classes only. The dataset used in our research provides seven lesions; these are melanocytic nevi, melanoma, benign keratosis, basal cell carcinoma, actinic keratoses, vascular lesions, and dermatofibroma. Thus, through the HAM10000 dataset and various deep learning models, we diversified the precision factors as well as input qualities. The obtained results are highly propitious and establish its credibility.

How insight contributes to diagnostic excellence

In the quest to improve diagnosis, a great deal of attention has already been focused on how to optimize clinical reasoning, and the importance of System 1 and System 2 processing. In this essay we consider the role of 'insight', a relatively overlooked pathway for arriving at the correct diagnosis. Insight refers to spontaneous emergence of the correct answer at some later point in time. We discuss factors that might facilitate insight, and how these could be incorporated into the diagnostic process.

Pandemic and the human factor

As the staffing crisis in the UK deepens, it is time for the policy-makers and professional bodies to rethink the approach to the most vital and yet most fragile component of the healthcare system-the human beings. The austerity measures, combined with pandemic and more recently the vision of a backlog with attached unrealistic expectations of tackling it, have brought the NHS and many other healthcare systems to the brink of a crisis. It is a human factors approach, which emphasises clinician's well-being as the core aspect of optimising performance that should become our goal. Delivery of healthcare under circumstances of physical, legal or moral threat cannot be optimal and is not sustainable. The pandemic served to highlight this quite clearly. Also, an injured, tired or burn-out healthcare professional cannot be expected to repair the system that has precipitated his or her condition. The approach to changing the culture of medicine may be multifaceted, but ultimately, we should rethink professionalism and the definition of duty of care putting emphasis on the well-being of those delivering the care as the way to assure best possible care.

Diagnostic journeys: characterization of patients and diagnostic outcomes from an academic second opinion clinic

OBJECTIVES

Diagnostic programs and second opinion clinics have grown and evolved in the recent years to help patients with rare, puzzling, and complex conditions who often suffer prolonged diagnostic journeys, but there is a paucity of literature on the clinical characteristics of these patients and the efficacy of these diagnostic programs. This study aims to characterize the diagnostic journey, case features, and diagnostic outcomes of patients referred to a team-based second opinion clinic at Stanford.

METHODS

Retrospective chart review was performed for 237 patients evaluated for diagnostic second opinion in the Stanford Consultative Medicine Clinic over a 5 year period. Descriptive case features and diagnostic outcomes were assessed, and correlation between the two was analyzed.

RESULTS

Sixty-three percent of our patients were women. 49% of patients had a potential precipitating event within about a month prior to the start of their illness, such as medication change, infection, or medical procedure. A single clear diagnosis was determined in 33% of cases, whereas the remaining cases were assessed to have multifactorial contributors/diagnoses (20%) or remained unclear despite extensive evaluation (47%). Shorter duration of illness, fewer prior specialties seen, and single chief symptom were associated with higher likelihood of achieving a single clear diagnosis.

CONCLUSIONS

A single-site academic consultative service can offer additional diagnostic insights for about half of all patients evaluated for puzzling conditions. Better understanding of the clinical patterns and patient experiences gained from this study helps inform strategies to shorten their diagnostic odysseys.

Pharmacists can improve diagnosis and help prevent diagnostic errors

We present two cases that highlight the role of pharmacists in the diagnostic process and illustrate how a culture of safety and teamwork between pharmacists and physicians can help prevent diagnostic errors.

System 2 Diagnostic Process for the Next Generation of Physicians: "Inside" and "Outside" Brain-The Interplay between Human and Machine

Improving diagnosis has been one of the most critical issues in medicine for the last two decades. In the context of the rise of digital health and its augmentation and human diagnostic thinking, it has become necessary to integrate the concept of digital diagnosis into dual-process theory (DPT), which is the fundamental axis of the diagnostic thinking process physicians. Particularly, since the clinical decision support system (CDSS) corresponds to analytical thinking (system 2) in DPT, it is necessary to redefine system 2 to include the CDSS. However, to the best of my knowledge there has been no concrete conceptual model based on this need. The innovation and novelty of this paper are that it redefines system 2 to include new concepts and shows the relationship among the breakdown of system 2. In this definition, system 2 is divided into “inside” and “outside” brains, where “inside” includes symptomatologic, anatomical, biomechanical–physiological, and etiological thinking approaches, and “outside” includes CDSS. Moreover, this paper discusses the actual and possible future interplay between “inside” and “outside.” The author envisions that this paper will serve as a cornerstone for the future development of system 2 diagnostic thinking strategy.

Clinical decision support in primary care for better diagnosis and management of retinal disease

Misdiagnosis of retinal disease is a common problem in primary care that can lead to irreversible vision loss and false-positive referrals, resulting in inappropriate use of health services. Clinical decision support systems describe tools that leverage information technology to provide timely recommendations that assist clinicians in the decisions they make about the care of a patient. They, therefore, have the potential to reduce the rate of misdiagnosis by promoting evidence-based medicine and more effective and efficient healthcare. This narrative review aims to support primary care practitioners in better understanding the current and emerging capacity of clinical decision support systems in eye care. Different types of clinical decision support systems are discussed, using current examples and evidence from the available literature to demonstrate how they may improve diagnostic effectiveness and aid the management of retinal disease. Comments are made on the future directions of clinical decision support in primary eye care and the potential applications of artificial intelligence.

Organization of Pediatric Echocardiography Laboratories: Impact of Sonographers on Clinical, Academic, and Financial Performance

Echocardiography has evolved the first-line imaging for diagnosis and management of pediatric and congenital heart disease all over the world. While it recognized as essential component of pediatric cardiac care delivery, organization of pediatric echocardiography services is very heterogeneous across the world, mainly related to significant differences in material and human resources in heterogeneous health care systems. In this paper, we focus on the role of pediatric sonographers, defined as expert technicians in pediatric echocardiography. While in some services sonographers are an essential part of the organizational structure, other laboratories operate only with physicians trained in echocardiography. The impact of sonographers on clinical, academic and financial performance will be discussed. Two organizational models (with and without sonographers) will be compared, and the advantages and disadvantages of each model will be evaluated. Different models of care provision are possible and decisions on organizational models need to be adjusted to the demands and available resources.

Toward a coherent structuration of disorders of consciousness expertise at a country scale: A proposal for France

Probing consciousness and cognitive abilities in non-communicating patients is one of the most challenging diagnostic issues. A fast growing medical and scientific literature explores the various facets of this challenge, often coined under the generic expression of 'Disorders of Consciousness' (DoC). Crucially, a set of independent converging results demonstrated both (1) the diagnostic and prognostic importance of this expertise, and (2) the need to combine behavioural measures with brain structure and activity data to improve diagnostic and prognostication accuracy as well as potential therapeutic intervention. Thus, probing consciousness in DoC patients appears as a crucial activity rich of human, medical, economic and ethical consequences, but this activity needs to be organized in order to offer this expertise to each concerned patient. More precisely, diagnosis of consciousness differs in difficulty across patients: while a minimal set of data can be sufficient to reach a confident result, some patients need a higher level of expertise that relies on additional behavioural and brain activity and brain structure measures. In order to enable this service on a systematic mode, we present two complementary proposals in the present article. First, we sketch a structuration of DoC expertise at a country-scale, namely France. More precisely, we suggest that a 2-tiers network composed of local (Tier-1) and regional (Tier-2) centers backed by distant electronic databases and algorithmic centers could optimally enable the systematic implementation of DoC expertise in France. Second, we propose to create a national common register of DoC patients in order to better monitor this activity, to improve its performance on the basis of nation-wide collected evidence, and to promote rational decision-making.

Lessons in clinical reasoning - pitfalls, myths, and pearls: the contribution of faulty data gathering and synthesis to diagnostic error

OBJECTIVES

Errors in clinical reasoning are a major factor for delayed or flawed diagnoses and put patient safety at risk. The diagnostic process is highly dependent on dynamic team factors, local hospital organization structure and culture, and cognitive factors. In everyday decision-making, physicians engage that challenge partly by relying on heuristics - subconscious mental short-cuts that are based on intuition and experience. Without structural corrective mechanisms, clinical judgement under time pressure creates space for harms resulting from systems and cognitive errors. Based on a case-example, we outline different pitfalls and provide strategies aimed at reducing diagnostic errors in health care.

CASE PRESENTATION

A 67-year-old male patient was referred to the neurology department by his primary-care physician with the diagnosis of exacerbation of known myasthenia gravis. He reported shortness of breath and generalized weakness, but no other symptoms. Diagnosis of respiratory distress due to a myasthenic crisis was made and immunosuppressive therapy and pyridostigmine were given and plasmapheresis was performed without clinical improvement. Two weeks into the hospital stay, the patient's dyspnea worsened. A CT scan revealed extensive segmental and subsegmental pulmonary emboli.

CONCLUSIONS

Faulty data gathering and flawed data synthesis are major drivers of diagnostic errors. While there is limited evidence for individual debiasing strategies, improving team factors and structural conditions can have substantial impact on the extent of diagnostic errors. Healthcare organizations should provide the structural supports to address errors and promote a constructive culture of patient safety.

The reduction of race and gender bias in clinical treatment recommendations using clinician peer networks in an experimental setting

Bias in clinical practice, in particular in relation to race and gender, is a persistent cause of healthcare disparities. We investigated the potential of a peer-network approach to reduce bias in medical treatment decisions within an experimental setting. We created "egalitarian" information exchange networks among practicing clinicians who provided recommendations for the clinical management of patient scenarios, presented via standardized patient videos of actors portraying patients with cardiac chest pain. The videos, which were standardized for relevant clinical factors, presented either a white male actor or Black female actor of similar age, wearing the same attire and in the same clinical setting, portraying a patient with clinically significant chest pain symptoms. We found significant disparities in the treatment recommendations given to the white male patient-actor and Black female patient-actor, which when translated into real clinical scenarios would result in the Black female patient being significantly more likely to receive unsafe undertreatment, rather than the guideline-recommended treatment. In the experimental control group, clinicians who were asked to independently reflect on the standardized patient videos did not show any significant reduction in bias. However, clinicians who exchanged real-time information in structured peer networks significantly improved their clinical accuracy and showed no bias in their final recommendations. The findings indicate that clinician network interventions might be used in healthcare settings to reduce significant disparities in patient treatment.

[Digital diagnostic support in rheumatology]

Only the correct diagnosis enables an effective treatment of rheumatic diseases. Digitalization has already significantly accelerated and simplified our everyday life. An increasing number of digital options are available to patients and medical personnel in rheumatology to accelerate and improve the diagnosis. This work gives an overview of current developments and tools for patients and rheumatologists, regarding digital diagnostic support in rheumatology.

Artificial Intelligence: Has Its Time Come for Inclusion in Medical School Education? Maybe…Maybe Not

This article was migrated. The article was marked as recommended. Artificial intelligence (AI) has the potential to strongly modify or even transform the landscape of medicine. Judicious utilization of AI can further propel progress in medical research, facilitate precision medicine, and optimize clinical workflow-the applications are limitless. Although technology and AI algorithms are evolving rapidly and have important implications for future physicians, there is a dearth of literature available for medical students and their educators to learn about AI. While scientific journals provide information regarding AI, they often are written for and by scientists, engineers, and physicians who are well-versed in technology. Currently, medical students must navigate the technical jargon and decipher AI literature without any foundational exposure. It is difficult for students to understand the implications of AI if they do not have basic knowledge of AI and its current capabilities. A fear about AI is pervasive amongst medical students. There is virtually no literature that provides a fundamental and easily digestible overview of AI for medical students and educators, while also offering suggestions about how to integrate AI into medical school curricula. Our goal is to address the lack of formal AI instruction by presenting an informative primer with curricular suggestions for each medical school year, from a U.S. perspective, tailored to medical students and their educators. We seek to present a balanced approach, as there are pros and cons about incorporating AI in undergraduate medical education.

Comparison of Diagnostic Recommendations from Individual Physicians versus the Collective Intelligence of Multiple Physicians in Ambulatory Cases Referred for Specialist Consultation

BACKGROUND

Studies report higher diagnostic accuracy using the collective intelligence (CI) of multiple clinicians compared with individual clinicians. However, the diagnostic process is iterative, and unexplored is the value of CI in improving clinical recommendations leading to a final diagnosis.

METHODS

To compare the appropriateness of diagnostic recommendations advised by individual physicians versus the CI of physicians, we entered actual consultation requests sent by primary care physicians to specialists onto a web-based CI platform capable of collecting diagnostic recommendations (next steps for care) from multiple physicians. We solicited responses to 35 cases (12 endocrinology, 13 gynecology, 10 neurology) from ≥3 physicians of any specialty through the CI platform, which aggregated responses into a CI output. The primary outcome was the appropriateness of individual physician recommendations versus the CI output recommendations, using recommendations agreed upon by 2 specialists in the same specialty as a gold standard. The secondary outcome was the recommendations' potential for harm.

RESULTS

A total of 177 physicians responded. Cases had a median of 7 respondents (interquartile range: 5-10). Diagnostic recommendations in the CI output achieved higher levels of appropriateness (69%) than recommendations from individual physicians (45%; χ² = 5.95, P = 0.015). Of the CI recommendations, 54% were potentially harmful, as compared with 41% of individuals' recommendations (χ² = 2.49, P = 0.11).

LIMITATIONS

Cases were from a single institution. CI was solicited using a single algorithm/platform.

CONCLUSIONS

When seeking specialist guidance, diagnostic recommendations from the CI of multiple physicians are more appropriate than recommendations from most individual physicians, measured against specialist recommendations. Although CI provides useful recommendations, some have potential for harm. Future research should explore how to use CI to improve diagnosis while limiting harm from inappropriate tests/therapies.

Diagnostic yield and accuracy in a tertiary referral syncope unit validating the ESC guideline on syncope: a prospective cohort study

Aims 

To assess in patients with transient loss of consciousness the diagnostic yield, accuracy, and safety of the structured approach as described in the ESC guidelines in a tertiary referral syncope unit.

Methods and results 

Prospective cohort study including 264 consecutive patients (≥18 years) referred with at least one self-reported episode of transient loss of consciousness and presenting to the syncope unit between October 2012 and February 2015. The study consisted of three phases: history taking (Phase 1), autonomic function tests (AFTs) (Phase 2), and after 1.5-year follow-up with assessment by a multidisciplinary committee (Phase 3). Diagnostic yield was assessed after Phases 1 and 2. Empirical diagnostic accuracy was measured for diagnoses according to the ESC guidelines after Phase 3. The diagnostic yield after Phase 1 (history taking) was 94.7% (95% CI: 91.1–97.0%, 250/264 patients) and increased to 97.0% (93.9–98.6%, 256/264 patients) after Phase 2. The overall diagnostic accuracy (as established in Phase 3) of the Phases 1 and 2 diagnoses was 90.6% (95% CI: 86.2–93.8%, 232/256 patients). No life-threatening conditions were missed. Three patients died, two unrelated to the cause of transient loss of consciousness, and one whom remained undiagnosed.

Conclusion 

A clinical work-up at a tertiary syncope unit using the ESC guidelines has a high diagnostic yield, accuracy, and safety. History taking (Phase 1) is the most important diagnostic tool. Autonomic function tests never changed the Phase 1 diagnosis but helped to increase the certainty of the Phase 1 diagnosis in many patients and yield additional diagnoses in patients who remained undiagnosed after Phase 1. Diagnoses were inaccurate in 9.4%, but no serious conditions were missed. This is adequate for clinical practice.