Natural language processing to automate a web-based model of care and modernize skin cancer multidisciplinary team meetings.

BACKGROUND: Cancer multidisciplinary team (MDT) meetings are under intense pressure to reform given the rapidly rising incidence of cancer and national mandates for protocolized streaming of cases. The aim of this study was to validate a natural language processing (NLP)-based web platform to automate evidence-based MDT decisions for skin cancer with basal cell carcinoma as a use case. METHODS: A novel and validated NLP information extraction model was used to extract perioperative tumour and surgical factors from histopathology reports. A web application with a bespoke application programming interface used data from this model to provide an automated clinical decision support system, mapped to national guidelines and generating a patient letter to communicate ongoing management. Performance was assessed against retrospectively derived recommendations by two independent and blinded expert clinicians. RESULTS: There were 893 patients (1045 lesions) used to internally validate the model. High accuracy was observed when compared against human predictions, with an overall value of 0.92. Across all classifiers the virtual skin MDT was highly specific (0.96), while sensitivity was lower (0.72). CONCLUSION: This study demonstrates the feasibility of a fully automated, virtual, web-based service model to host the skin MDT with good system performance. This platform could be used to support clinical decision-making during MDTs as 'human in the loop' approach to aid protocolized streaming. Future prospective studies are needed to validate the model in tumour types where guidelines are more complex.

Service implications of the revised 2022 National Institute for Health and Care Excellence (NICE) follow-up guidelines for stage IA-IIC melanoma.

BACKGROUND: The 2022 National Institute for Health and Care Excellence melanoma guideline update made significant changes to follow-up. The aim of this study was to assess the impact these changes will have on a national melanoma cohort over a 5-year follow-up interval. METHODS: Anonymized, individual-level, population-scale, linkable primary and secondary care National Health Service data for an 18-year interval (2000-2018) in Wales, UK were analysed. These data were used to predict the number of patients over a 10-year interval (2020-2030) that would be diagnosed with melanoma. Follow-up schedules for the 2015 and 2022 National Institute for Health and Care Excellence melanoma guidelines were then used to calculate the number of clinician-led appointments, the number of radiological investigations, and the total healthcare cost between 2025 and 2030, corresponding to a 5-year patient follow-up interval, for those with stage IA-IIC melanoma. RESULTS: Between 2025 and 2030 it is predicted that implementation of the 2022 guidelines would lead to 21 122 (range 19 194-23 083) fewer clinician-led appointments for patients with stage IA-IIC melanoma. However, there would be a significant increase in the number of radiological investigations (7812; range 7444-8189). These changes would lead to a €2.74 million (€1.87 million-€3.61 million) reduction in the total cost of follow-up over the interval 2025-2030. CONCLUSION: Melanoma follow-up guideline changes will result in a substantial reduction in the number of clinical follow-up appointments, but a significant additional burden to radiological services. The overall cost of follow-up at a national level will be reduced.

Prevalence and odds of anxiety and depression in cutaneous malignant melanoma: a proportional meta-analysis and regression.

BACKGROUND: The psychological burden of cutaneous malignant melanoma (CM) is all-encompassing, affecting treatment adherence, recurrence and mortality. However, the prevalence and risk factors of anxiety and depression in CM remain unclear. OBJECTIVES: To establish a benchmark pooled prevalence of anxiety and depression in CM, to provide magnitudes of association for clinical, therapeutic and demographic correlates, and to elucidate temporal trends in anxiety and depression from the time of diagnosis. METHODS: This review followed the MOOSE guidelines. MEDLINE, Embase, PsychINFO, Web of Science and the Cochrane Library were queried from database inception to 24 August 2023. Study selection, data extraction and quality assessment were performed by two independent authors, utilizing both the Joanna Briggs Institute (JBI) and National Institutes of Health risk-of-bias tools for the latter. The GRADE approach was used to rate the certainty of evidence. Prevalence rates, 95% confidence intervals (CIs) and prediction intervals (PIs) were derived using a random-effects model and estimating between- and within-study variance. RESULTS: Nine longitudinal and 29 cross-sectional studies were included (7995 patients). Based on the JBI and NIH tools, respectively, quality assessment found 20 and 17 to be at low risk of bias, 12 and 15 to be at moderate risk and 6 and 5 to be at high risk of bias. The prevalence of anxiety [30.6% (95% CI 24.6-37.0; PI 18-47%)] and depression [18.4% (95% CI 13.4-23.9; PI 10-33%)] peaked during treatment, declining to pretreatment levels after 1 year [anxiety: 48% vs. 20% (P = 0.005); depression: 28% vs. 13% (P = 0.03)]. Female sex [odds ratio (OR) 1.8, 95% CI 1.4-2.3; P < 0.001], age < 60 years (OR 1.5, 95% CI 1.2-2.0; P = 0.002) and low educational level (OR 1.5, 95% CI 1.2-2.0; P < 0.001) were likely to result in a large increase in the odds of anxiety. Depression was 12.3% higher in those with stage IV vs. those with stage I CM (P = 0.05). Relative to immune checkpoint inhibition, the rates of depression were 22% (P = 0.002) and 34% (P < 0.001) higher among patients with advanced-stage CM receiving interferon-α and chemotherapy, respectively. A significant reduction in self-reported depression scores was demonstrated over time (P = 0.003). CONCLUSIONS: Notably, anxiety and depression in CM affect women, those younger than 60 years of age and the less educated, with up to 80% higher odds of anxiety in these groups. Anxiety and depression surge during chemotherapy and interferon treatment, especially in advanced CM. Our findings facilitate risk stratification and underscore the need for multidisciplinary vigilance.

Melanoma is a serious type of skin cancer that is becoming more prevalent, particularly in people with lighter skin. The UK-based ReconRegen research group conducted a study to understand the psychological impact of melanoma on people, focusing on anxiety and depression. To do this, a systematic review approach was used to analyse data from existing studies and gather a comprehensive perspective. The study discovered that 30% of people with melanoma are affected by anxiety and 18% by depression, significantly higher than the general population. Key risk factors for anxiety included being female, being younger than 60 years of age and having lower educational attainment. Women are 1.8 times more likely to experience anxiety than men, those under 60 years of age are 1.5 times more likely to experience it and individuals with lower educational levels are also 1.5 times more likely to experience anxiety. Findings showed that anxiety and depression levels peaked during treatment phases, especially in people undergoing chemotherapy and immunotherapy. This highlights the need for targeted mental health support during these treatment periods. The findings advocate for mental health considerations in melanoma care, suggesting regular mental health assessments, particularly for high-risk groups and during intense treatment phases. Highlighting the importance of a holistic treatment approach, the study suggests that future research should include long-term studies to understand the chronic impacts of anxiety and depression. Improved clarity and detail in research reporting are essential for developing effective mental health support for people with melanoma, enhancing overall patient care by addressing both physical and emotional health needs.

The 'Flapbot'- a global perspective on the validity and usability of a flap monitoring chatbot.

BACKGROUND: The Flapbot chatbot assists in free flap monitoring, emphasising accessibility, user-friendliness, and global reliability. This study assesses Flapbot's worldwide validity and usability and uses qualitative analysis to identify areas for future enhancement. METHODS: Flapbot, built on Google's Dialogflow, was evaluated by international plastic surgeons. Invitations were sent to the International Lower Limb Reconstruction Collaborative (INTELLECT), International Confederation of Plastic Surgery Societies (ICOPLAST), and the International Microsurgery Club. Out of the 42 surgeons who agreed to participate 21 tested the Flapbot and completed an online survey on its validity and usability. The survey had 13 validity items and 10 usability items. Data analysis involved computing the Individual Content Validity Index (I-CVI) and Scale-wide Content Validity Index (S-CVI) for validity, and the System Usability Score (SUS) for usability. Thematic analysis distilled free text responses to identify key themes. RESULTS: Nine of thirteen items had an I-CVI over 0.78, denoting significant relevance. The S-CVI score stood at 0.82, indicating high relevance. The SUS score was 68, representing average usability. Themes highlighted issues with the current model, development suggestions, and surgeons' concerns regarding growing reliance on digital tools in healthcare. CONCLUSION: Flapbot is a promising digital aid for free flap monitoring. While it showcases notable validity and usability, improvements in functionality, usability, and accessibility are needed for broader global use.

Standardising the management of open extremity fractures: a scoping review of national guidelines.

INTRODUCTION: Open extremity fractures can be life-changing events. Clinical guidelines on the management of these injuries aim to standardise the care of patients by presenting evidence-based recommendations. We performed a scoping systematic review to identify all national clinical practice guidelines published to date. MATERIALS AND METHODS: A PRISMA-compliant scoping systematic review was designed to identify all national or federal guidelines for the management of open fractures, with no limitations for language or publication date. EMBASE and MEDLINE database were searched. Article screening and full-text review was performed in a blinded fashion in parallel by two authors. RESULTS: Following elimination of duplicates, 376 individual publications were identified and reviewed. In total, 12 clinical guidelines were identified, authored by groups in the UK, USA, the Netherlands, Finland, and Malawi. Two of these focused exclusively on antibiotic prophylaxis and one on combat-related injuries, with the remaining nine presented wide-scope recommendations with significant content overlap. DISCUSSION: Clinical practice guidelines serve clinicians in providing evidence-based and cost-effective care. We only identified one open fractures guideline developed in a low- or middle-income country, from Malawi. Even though the development of these guidelines can be time and resource intensive, the benefits may outweigh the costs by standardising the care offered to patients in different healthcare settings. International collaboration may be an alternative for adapting guidelines to match local resources and healthcare systems for use across national borders.

Critical Appraisal of the Quality of Evidence Addressing the Diagnosis, Prognosis, and Management of Peripheral Artery Disease in Patients With Diabetic Foot Ulceration.

AIMS: There is a paucity of robust evidence on prevention and management of diabetic foot ulcers (DFUs) to inform treatment. This study appraises the current quality of the evidence addressing diagnosis, prognosis, and management of peripheral artery disease (PAD) in patients with DFUs using a newly devised 21 point scoring (TOPS) disease specific research appraisal tool published by the International Working Group on the Diabetic Foot (IWGDF) and European Wound Management Association. METHODS: The 2015 IWGDF guidance on diagnosis, prognosis, and management of PAD in patients with DFUs was used to identify studies pertaining to prevention and management. Two reviewers assessed these articles against TOPS, which examines study design, conduct, and outcome reporting. RESULTS: The overall median score was 8 (3-12/21). The median design total score was 2 (0-4/11). The median conduct total score was 2 (1-4/6). The median outcomes total score was 3 (1-4/4). There was improvement with time in overall total (Spearman Rho 0.39, p = .0005), design total (0.35, p = .0023), and outcomes total (0.35, p = .0002), but not conduct total (-0.03, p = .8132) scores. CONCLUSIONS: Although this analysis revealed an improvement over time in the overall calibre of studies, the present quality remains poor on which to inform the diagnosis, management, and prognosis of patients with PAD and diabetic foot ulceration.

Can Large Language Models Generate Outpatient Clinic Letters at First Consultation That Incorporate Complication Profiles From UK and USA Aesthetic Plastic Surgery Associations?

The importance of written communication between clinicians and patients, especially in the wake of the Supreme Court case of Montgomery vs Lanarkshire, has led to a shift toward patient-centric care in the United Kingdom. This study investigates the use of large language models (LLMs) like ChatGPT and Google Bard in enhancing clinic letters with gold-standard complication profiles, aiming to improve patients' understanding and save clinicians' time in aesthetic plastic surgery. The aim of this study is to assess the effectiveness of LLMs in integrating complication profiles from authoritative sources into clinic letters, thus enhancing patient comprehension and clinician efficiency in aesthetic plastic surgery. Seven widely performed aesthetic procedures were chosen, and complication profiles were sourced from the British Association of Aesthetic Plastic Surgeons (BAAPS) and the American Society of Plastic Surgeons (ASPS). We evaluated the proficiency of the ChatGPT4, ChatGPT3.5, and Google Bard in generating clinic letters which incorporated complication profiles from online resources. These letters were assessed for readability using an online tool, targeting a recommended sixth-grade reading level. ChatGPT4 achieved the highest compliance in integrating complication profiles from BAAPS and ASPS websites, with average readability grades between eighth and ninth. ChatGPT3.5 and Google Bard showed lower compliance, particularly when accessing paywalled content like the ASPS Informed Consent Bundle. In conclusion, LLMs, particularly ChatGPT4, show promise in enhancing patient communications in aesthetic plastic surgery by effectively incorporating standard complication profiles into clinic letters. This aids in informed decision making and time saving for clinicians. However, the study underscores the need for improvements in data accessibility, search capabilities, and ethical considerations for optimal LLM integration into healthcare communications. Future enhancements should focus on better interpretation of inaccessible formats and a Human in the Loop approach to combine Artifical Intelligence capabilities with clinician expertise.

Revisiting basal cell carcinoma clinical margins: Leveraging natural language processing and multivariate analysis with updated Royal College of Pathologists histological reporting standards.

INTRODUCTION: Data supporting the current British Association of Dermatologists guidelines for the management of basal cell carcinoma (BCC) are based on historic studies and do not consider the updated Royal College of Pathologists (RCPath) histological reporting standards. The aim of this study was to use natural language processing (NLP)-derived data and undertake a multivariate analysis with updated RCPath standards, providing a contemporary update on the excision margins required to achieve histological clearance in BCC. METHODS: A validated NLP information extraction model was used to perform a rapid multi-centre, pan-specialty, consecutive retrospective analysis of BCCs, managed with surgical excision using a pre-determined clinical margin, over a 17-year period (2004-2021) at Swansea Bay University Health Board. Logistic regression assessed the relationship between the peripheral and deep margins and histological clearance. RESULTS: We ran our NLP algorithm on 34,955 BCCs. Out of the 1447 BCCs that met the inclusion criteria, the peripheral margin clearance was not influenced by the BCC risk level (p = 0.670). A clinical peripheral margin of 6 mm achieved a 95% histological clearance rate (95% confidence interval [CI], 0.93-0.98). Tumour thickness inversely affected deep-margin histological clearance (OR 0.720, 95% CI, 0.525-0.991, p < 0.05). Depth level 2 had a 97% probability of achieving deep-margin histological clearance across all tumour thicknesses. CONCLUSION: Updated RCPath reporting standards minimally impact the peripheral margin histological clearance in BCC. Larger clinical peripheral margins than those indicated by current guidelines may be necessary to achieve excision rates of ≥95%. These findings emphasise the need for continuous reassessment of clinical standards to enhance patient care.

An illustrated anatomical approach to reducing vascular risk during facial soft tissue filler administration - a review.

Vascular complications from soft tissue fillers can have catastrophic consequences for patients. Adverse events are rare, but they are increasing, and their appearance may be the result of intravascular injection. A comprehensive understanding of the 2-dimensional anatomy (distribution) and 3-dimensional anatomy (depth) of the facial vasculature is fundamental for the safe delivery of nonsurgical cosmetic procedures. The purpose of this review is to provide an illustrated approach to examine surgical anatomy specific to the facial vascular system and the anatomical considerations clinicians need to give in specific danger during injectable cosmetic procedures. A grounding in safety and anatomy will help the new injector to mitigate the risk of vascular complications.

Models and materials for teaching auricular framework carving: A systematic review.

INTRODUCTION: The process of carving an auricular framework is technically challenging and unique to the patient. As such, there is a need for a robust and reliable training model for practicing and planning ear reconstruction. The aim of this study is to assess the best models and methods available to practice the carving of an auricular framework. METHODS: A systematic review was undertaken in accordance with Preferred Reporting Items for Systematic Reviews and Meta Analyses guidelines using MEDLINE, Embase, and Cochrane databases. Terms such as "ear", "reconstruction" and "teaching" were searched. RESULTS: A total of 354 articles were identified, and 13 studies met the inclusion criteria. Vegetables, animal tissue, synthetic materials, as well as more advanced methods such as 3D-printed moulds, were analysed. The similarity of these materials to human costal cartilage was investigated to determine the best or most suitable ones for the purpose. The methods used in the studies were also analysed. Due to heterogeneity of the studies, it was not possible to conduct a quantitative analysis. CONCLUSION: This review identifies that for the junior surgeons at the skill acquisition phase of their training in auricular framework carving repetition using firstly a cheap synthetic material would be most useful, followed by animal cartilage or 3D printing using silicone. These materials bear the most resemblance to human costal cartilage, and by repeating the carvings, proficiency will improve. Those surgeons with an established ear reconstruction practise, wishing to undertake immediate pre-operatively simulation could benefit from cross-sectional imaging and 3D printing of a patient's non-affected ear to ensure a good match.

Improving the effectiveness of multidisciplinary team meetings on skin cancer: Analysis of the National Cancer Research UK survey responses.

INTRODUCTION: Skin cancer is the most common form of cancer in the UK, comprising at least 25% of all new cancer diagnoses. Many patients require referral to the local or specialist skin cancer multidisciplinary team (MDT) for ongoing management. However, national data have shown that Specialist Skin Cancer MDTs are costly and do not currently meet NICE standards for composition and quoracy. Innovative solutions to these problems are therefore warranted. METHODS: We performed a secondary comparative analysis of 3563 quantitative responses to two Cancer Research UK commissioned surveys along with subanalysis of 282 skin cancer MDT respondents. RESULTS: Good uniformity was observed amongst skin respondents in the belief that risk stratification and prioritization of complex cases were the most important factors compared to other cancer MDT members. The most important priorities for areas requiring change to MDT working deemed by the skin MDT were 1) imaging and pathology results ready for the meeting, 2) time to discuss patients in detail, 3) clear meeting owner in charge, and 4) clear agenda, in advance of the meeting. There was agreement (median Likert score 4) amongst skin MDT respondents that patients should be placed on protocolized treatment pathways. CONCLUSION: The responses of skin MDT respondents analyzed in the current study support changes to meeting attendance, preparation, and protocolized streaming. In line with other studies, we support tumor-specific guidance for streamlining MDT discussions. We also encourage stakeholders to adopt an evidence-based approach to test, develop, and reassess changes in this herculean task.

Evaluating remote skin cancer multidisciplinary team meetings in the United Kingdom post-COVID-19.

INTRODUCTION: Prior to the COVID-19 pandemic, there was concern that virtual or remote multidisciplinary teams (MDT) meetings represented a niche concept that was unlikely to replace traditional face-to-face meetings in the management of cancer. However, the sudden shift to virtual meetings during COVID-19 has been one of the most dramatic changes since the inception of the MDT. This study aims to investigate the effectiveness of virtual skin MDTs since the move to virtual meetings. METHODS: A cross-sectional survey was sent to all Specialist Skin Cancer MDTs (SSMDTs) and the British Association of Plastic, Reconstructive, and Aesthetic Surgeons Skin Oncology Special Interest and Advisory Group. RESULTS: There were 68 responses (55.3% response rate) from 36 SSMDTs in the UK. Respondents felt communication, chairing, and decision-making were similar in virtual and in-person MDTs, but the team working was worse in virtual meetings. Recruitment, data security, and patient confidentiality were maintained in virtual MDTs. Most preferred a hybrid format for future MDTs, with the option to attend virtually. Recommendations for improvement included better connectivity, IT support, training, and staff integration. CONCLUSION: The virtual MDT is here to stay. We highlight the strengths and weaknesses of remote virtual skin MDTs. It is key that we look at ways to retain team working to ensure that the collegiate nature of MDT working, and therefore treatment options for patients, are not lost in this transformation in MDT delivery.

Comparative study of ChatGPT and human evaluators on the assessment of medical literature according to recognised reporting standards.

INTRODUCTION: Amid clinicians' challenges in staying updated with medical research, artificial intelligence (AI) tools like the large language model (LLM) ChatGPT could automate appraisal of research quality, saving time and reducing bias. This study compares the proficiency of ChatGPT3 against human evaluation in scoring abstracts to determine its potential as a tool for evidence synthesis. METHODS: We compared ChatGPT's scoring of implant dentistry abstracts with human evaluators using the Consolidated Standards of Reporting Trials for Abstracts reporting standards checklist, yielding an overall compliance score (OCS). Bland-Altman analysis assessed agreement between human and AI-generated OCS percentages. Additional error analysis included mean difference of OCS subscores, Welch's t-test and Pearson's correlation coefficient. RESULTS: Bland-Altman analysis showed a mean difference of 4.92% (95% CI 0.62%, 0.37%) in OCS between human evaluation and ChatGPT. Error analysis displayed small mean differences in most domains, with the highest in 'conclusion' (0.764 (95% CI 0.186, 0.280)) and the lowest in 'blinding' (0.034 (95% CI 0.818, 0.895)). The strongest correlations between were in 'harms' (r=0.32, p<0.001) and 'trial registration' (r=0.34, p=0.002), whereas the weakest were in 'intervention' (r=0.02, p<0.001) and 'objective' (r=0.06, p<0.001). CONCLUSION: LLMs like ChatGPT can help automate appraisal of medical literature, aiding in the identification of accurately reported research. Possible applications of ChatGPT include integration within medical databases for abstract evaluation. Current limitations include the token limit, restricting its usage to abstracts. As AI technology advances, future versions like GPT4 could offer more reliable, comprehensive evaluations, enhancing the identification of high-quality research and potentially improving patient outcomes.

The incidence of non-melanoma skin cancer in the UK and the Republic of Ireland: a systematic review.

BACKGROUND: Basal cell carcinoma (BCC) and cutaneous squamous cell carcinoma (cSCC) remain the most prevalent malignancies, contributing a higher workload to cancer registries than all cancers combined. The nature of skin cancers in addition to current coding methods employed by registries give a skewed representation of the workload. OBJECTIVES: A comprehensive search examining the incidence of BCC and/or cSCC at a regional or national level in the UK and Ireland was devised. Standardisation methods were grouped to permit comparison, and a bias assessment tool was employed. MATERIALS & METHODS: Sixteen UK and Republic of Ireland epidemiological studies on BCC/cSCC over a 48-year period were compiled, examining incidence, trends and emerging risk factors. RESULTS: Incidence of BCC within the UK is increasing annually by up to 4%, with rates in Wales increasing by up to 6.6% and 1.6% annually for BCC and cSCC, respectively. Inverse relationships are noted between BCC/cSCC and social deprivation. Although the elderly remain the most at risk, the 30-49 age group have illustrated growth rates of approximately 4%. CONCLUSION: This review outlines increasing incidence in the UK with higher rates noted in Wales, the Southeast, West and Central regions. Incidence rates are higher amongst the least socially deprived and an increasing incidence amongst younger age groups was found, however further trend analysis is required. A more comprehensive data collection method within registries is necessary to ensure accurate representation and fluid comparison. Service planning and public awareness campaigns must be implemented to prevent overwhelming future services.

Using a ChatBot to support clinical decision-making in free flap monitoring.

Flap monitoring charts and escalation protocols are ubiquitous amongst microsurgical departments and can facilitate converting flap observations into flap monitoring decisions. However, human factors in the recognition-communication process of decision-making still pose a threat to timely intervention and thus are a key determinant of success in microvascular surgery. Digitally transforming paper-based pathways may facilitate early recognition and escalation to potentially salvage a free flap with compromised vascularity. We describe the early adoption and use case of a ChatBot to support clinical decision-making support for free flap monitoring - the 'FlapBot'.

Diagnosing and managing a giant primary cutaneous malignant melanoma of the lower limb.

We present a woman who was referred to our plastic surgery unit with a suspected squamous cell carcinoma following a 3-year history of an enlarging mass on her thigh. Surprisingly, histopathological assessment confirmed the diagnosis of nodular malignant melanoma measuring 77×77×54 mm with a Breslow thickness of 52 mm, making it the largest recorded lower limb primary cutaneous malignant melanoma in the UK.