Medical Experts Research Articles

Apologizing for Intergroup Criticism Reduces Rejection of Public Health Officials’ Pro-Vaccine Messages

ABSTRACT Compared to other medical experts, public health officials face added scrutiny of their pro-vaccine messages due to distrust of the US federal government. We consider reactions to such critical messages through the lens of conversational norms. Conversational pragmatics suggest that polite communication is essentially cooperative in nature, avoiding criticism of the other or excessive praise of the self. Applied to intergroup communication, this suggests that critical messages from outgroups will be viewed as impolite. Distrusted outgroup sources (such as government officials) are particularly expected to be uncooperative and impolite, increasing message rejection. Importantly, this perspective suggests that apologizing for the norm violation should mitigate these effects. Applying this reasoning to scientific messaging, we demonstrate that a message critical of the under-vaccinated is more likely to be rejected if made by a US federal public health official than by a private medical expert. However, this effect is reduced if the public health official apologizes in advance for violating the norm against criticizing other groups. Improved perceptions of the apologizing source’s motives and reduced message threat were associated with higher vaccination intentions. Practical and theoretical implications for intergroup communication are discussed.

Bridging healthcare gaps: a scoping review on the role of artificial intelligence, deep learning, and large language models in alleviating problems in medical deserts.

"Medical deserts" are areas with low healthcare service levels, challenging the access, quality, and sustainability of care. This qualitative narrative review examines how artificial intelligence (AI), particularly large language models (LLMs), can address these challenges by integrating with e-Health and the Internet of Medical Things to enhance services in under-resourced areas. It explores AI-driven telehealth platforms that overcome language and cultural barriers, increasing accessibility. The utility of LLMs in providing diagnostic assistance where specialist deficits exist is highlighted, demonstrating AI's role in supplementing medical expertise and improving outcomes. Additionally, the development of AI chatbots offers preliminary medical advice, serving as initial contact points in remote areas. The review also discusses AI's role in enhancing medical education and training, supporting the professional development of healthcare workers in these regions. It assesses AI's strategic use in data analysis for effective resource allocation, identifying healthcare provision gaps. AI, especially LLMs, is seen as a promising solution for bridging healthcare gaps in "medical deserts," improving service accessibility, quality, and distribution. However, continued research and development are essential to fully realize AI's potential in addressing the challenges of medical deserts.

Implementasi K-Means Clustering pada Pengelompokan Pasien Penyakit Jantung

Heart disease is a prominent global health concern, necessitating early identification and patient grouping for effective management. This study employs the K-Means clustering algorithm with a medical dataset of 303 patients, encompassing various attributes. These include Age, Gender, Chest Pain Type, Blood Pressure, Serum Cholesterol Level, Fasting Blood Sugar, Resting Electrocardiographic Results, Maximum Heart Rate, Angina, ST Depression, and Slope of the ST Segment. The goal is to categorize patients into four clusters based on chest pain types, a crucial symptom indicating disease severity. The computation concludes after the sixth iteration, revealing Cluster 1 (27 patients), Cluster 2 (135 patients), Cluster 3 (15 patients), and Cluster 4 (126 patients). Collaborative analysis with medical experts highlights that Cluster 1, mainly comprising older males, exhibits high-risk indicators. While this grouping aids in personalized treatment strategy development, further clinical validation involving more experts and datasets is imperative for enhanced reliability.

Clinical Clerkship With or Without Scheduled Patient Consultations: Does It Make a Difference to Medical Students’ Experiences of Learning?

Abstract Background Becoming a medical expert involves leadership and professionalism, which are critical skills to learn in medical education. However, a gap exists in understanding how didactic variations in the organisation of clinical clerkships impact medical students’ opportunities to develop these skills. This study explored how clinical clerkships with or without scheduled patient consultations affect medical students’ experiences of learning leadership and professional behaviour. Materials and Methods We conducted a qualitative, quasi-experimental study. Data were gathered through 11 end-of-clerkship focus group interviews with 87 fifth-year medical students who participated in one of two 8-week clerkships at a surgical department: a clerkship with a mentor physician or a clerkship with scheduled patient consultations in combination with a mentor physician. Using a constructivist lens, we analysed the focus group interview transcripts and applied grounded theory principles to the iterative coding process. Results The analyses resulted in a descriptive framework displaying nine themes. Each theme was described as a spectrum of meanings that offers a nuanced understanding of how didactic variations in the organisation of clerkship impact how students ascribed meaning to the learning experiences in undergraduate medical education, focusing on learning leadership and professional behaviour. Discussion This study investigates differences in student learning experiences when participating in a clerkship with or without scheduled patient consultation and, therefore, adds new insight into the need for a balanced, thoughtful, and didactic approach to medical education that considers both clinical exposure and co-regulated learning.

Включенность некоммерческих организаций и объединений в деятельность общественных комиссий при главных бюро медико-социальной экспертизы в субъектах Российской Федерации

The system of medical and social expertise in Russia is three years old. Involvement of public and the non-profit sector have resulted in a significant number of representatives of specialized non-profit organizations in public commissions under the chief bureaus of sociomedical assessment. Patient-oriented non-profit organizations introduce new guidelines into the system, e.g., a more rapid adjustment of medical and social expertise to the needs of the target group. Nevertheless, the quality and accessibility of sociomedical assessment services remain a prospective research field. The article describes the involvement of non-profit organizations and associations in the activities of public commissions under the chief bureaus of medical and social assessment. The research was part of the project on Public participation in improving the sociomedical assessment in the Russian Federation: involvement, development, and monitoring. The research data were obtained from structured interviews with representatives of public commissions at the chief bureaus of sociomedical assessment and focus group interviews. Two groups included 12–15 representatives of patient non-profit organizations involved in public commissions. The resulting analysis of the composition of public commissions for 2018–2022 revealed the following facts. Patient non-profit organizations were the most significant group among public commissions are representatives of patient non-profit organizations (34.1%). Representatives of the chief bureaus of sociomedical assessment tended to leave: 12% in 2018 and 4% in 2022. The wide representation of non-profit organizations indicated the public orientation of the commissions. The most active participants were those non-profit organizations that consulted citizens, thus improving the effectiveness of the commissions. The factors that affected the activity of public commissions were connected with their employees, i.e., their awareness, goal understanding, motivation for public work, involvement, and workflow management. Patient non-profit organizations participated in the work of public commissions in following types: presence, representation, and goal achievement. The current development vector is the shift to the type of goal achievement.

Therapeutic Role of Probiotics Against Environmental-Induced Hepatotoxicity: Mechanisms, Clinical Perspectives, Limitations, and Future.

Hepatotoxicity is one of the biggest health challenges, particularly in the context of liver diseases, often aggravated by gut microbiota dysbiosis. The gut-liver axis has been regarded as a key idea in liver health. It indicates that changes in gut flora caused by various hepatotoxicants, including alcoholism, acetaminophen, carbon tetrachloride, and thioacetamide, can affect the balance of the gut's microflora, which may lead to increased dysbiosis and intestinal permeability. As a result, bacterial endotoxins would eventually enter the bloodstream and liver, causing hepatotoxicity and inducing inflammatory reactions. Many treatments, including liver transplantation and modern drugs, can be used to address these issues. However, because of the many side effects of these approaches, scientists and medical experts are still hoping for a therapeutic approach with fewer side effects and more positive results. Thus, probiotics have become well-known as an adjunctive strategy for managing, preventing, or reducing hepatotoxicity in treating liver injury. By altering the gut microbiota, probiotics offer a secure, non-invasive, and economical way to improve liver health in the treatment of hepatotoxicity. Through various mechanisms such as regulation of gut microbiota, reduction of pathogenic overgrowth, suppression of inflammatory mediators, modification of hepatic lipid metabolism, improvement in the performance of the epithelial barrier of the gut, antioxidative effects, and modulation of mucosal immunity, probiotics play their role in the treatment and prevention of hepatotoxicity. This review highlights the mechanistic effects of probiotics in environmental toxicants-induced hepatotoxicity and current findings on this therapeutic approach's experimental and clinical trials.

Debates over the role of Traditional Chinese Medicine on COVID-19: A computational comparison between professionals and laypersons in Chinese online knowledge community

Leveraging a large collection of textual data (N = 21,539) from a Chinese online community, we employed structural topic modeling to investigate the thematic disparities between professionals and laypersons, regarding the effectiveness of Traditional Chinese Medicine (TCM) on COVID-19. Findings reveal that laypersons are the dominant communicators in terms of discussion volume, who often focus on relevant news events, societal or political aspects of TCM. In contrast, professionals keep concentrating on issues related to medical expertise, and do not shift attentions as frequent as laypersons. Despite the dominant influence of professionals on laypersons’ agenda, two-way agenda interactions identified confirm that lay public is empowered to negotiate with elite professionals under certain topics. Our results provide novel insights into the dynamic nature of attentions, behaviors, and relations among prominent communication actors, and encourage future research to examine the individual-level and societal-level impacts of these constructs in the emerging online media landscape.

Design of a Customized Thumb Prosthesis: Focus on Functionality and Comfort

This research focuses on addressing the scarcity of personalized and affordable prostheses in Honduras by designing and developing a functional prosthesis for a thumb. The general objective was to design a model suitable for 3D printing adapted to the specific needs of a patient who had her index and thumb fingers amputated. To achieve this, an integral methodology was employed, ranging from an exhaustive literature review of current solutions to detailed material evaluations, including interviews with patients and medical experts, SolidWorks modeling, and simulations. The result was the successful design of a functional prosthesis prototype, with the selection of intentionally 3D printable materials to ensure comfort and functionality. In conclusion, this research demonstrates the viability of addressing the scarcity of personalized prostheses through innovative approaches and comprehensive design methodologies, with a focus on effectiveness and streamlining the process.

Evaluating the Bias in Hospital Data: Automatic Preprocessing of Patient Pathways Algorithm Development and Validation Study.

The optimization of patient care pathways is crucial for hospital managers in the context of a scarcity of medical resources. Assuming unlimited capacities, the pathway of a patient would only be governed by pure medical logic to meet at best the patient's needs. However, logistical limitations (eg, resources such as inpatient beds) are often associated with delayed treatments and may ultimately affect patient pathways. This is especially true for unscheduled patients-when a patient in the emergency department needs to be admitted to another medical unit without disturbing the flow of planned hospitalizations. In this study, we proposed a new framework to automatically detect activities in patient pathways that may be unrelated to patients' needs but rather induced by logistical limitations. The scientific contribution lies in a method that transforms a database of historical pathways with bias into 2 databases: a labeled pathway database where each activity is labeled as relevant (related to a patient's needs) or irrelevant (induced by logistical limitations) and a corrected pathway database where each activity corresponds to the activity that would occur assuming unlimited resources. The labeling algorithm was assessed through medical expertise. In total, 2 case studies quantified the impact of our method of preprocessing health care data using process mining and discrete event simulation. Focusing on unscheduled patient pathways, we collected data covering 12 months of activity at the Groupe Hospitalier Bretagne Sud in France. Our algorithm had 87% accuracy and demonstrated its usefulness for preprocessing traces and obtaining a clean database. The 2 case studies showed the importance of our preprocessing step before any analysis. The process graphs of the processed data had, on average, 40% (SD 10%) fewer variants than the raw data. The simulation revealed that 30% of the medical units had >1 bed difference in capacity between the processed and raw data. Patient pathway data reflect the actual activity of hospitals that is governed by medical requirements and logistical limitations. Before using these data, these limitations should be identified and corrected. We anticipate that our approach can be generalized to obtain unbiased analyses of patient pathways for other hospitals.

Abstract B103: Friend-to-friend: Culturally tailored health education for East African communities

Abstract Purpose: The Mayo Clinic Comprehensive Cancer Center (MCCCC) serves its communities through every pillar of the cancer continuum from prevention to survivorship. To effectively educate our communities along this continuum, MCCCC partnered with a trusted community partner, leveraged the stories of survivors, and addressed stigma in a safe space for community members to ask vulnerable questions. Methods: The MCCCC partnered with Pamoja Women on the Friend-to-Friend initiative. Pamoja Women is a trusted community partner supporting East African women and their families, some of whom are refugees. They provided the space to hold each event, an environment where the women were comfortable with support from their community. The program highlighted cancer education, screening, and risk reduction through culturally appropriate discussions between friends. Topics were identified based on community interest and the cancer incidence in the community. Survivors shared their personal stories and medical experts discussed prevention, risk reduction, screening, addressed myths, answered questions, and showcased discoveries in research and treatment methods. Results: We have held 3 events covering breast, cervical, and HPV vaccination. Each event saw an increase in participation with almost 50 participants attending the series. Furthermore, there was an increase in awareness of screening methods, prevention, and risk reduction. Pamoja Women provided navigation to screening through transportation to and advocacy during appointments. This continuity from education to screening led to heading the call to action for screening. Due to its success, the four-part series has been expanded to include the topics of endometrial, prostate cancer, and breast cancer for a second time. Conclusion: Partnering with trusted community organizations has led to safe spaces for cancer education with friends. This approach has led to successful education and navigation to screening in communities that do not always have access to this essential information. Citation Format: Adeline Abbenyi, Manal Abbadi, Khadija Ali, Fatuma Ahmed. Friend-to-friend: Culturally tailored health education for East African communities [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr B103.

In Silico Decoding of Parkinson's: Speech & Writing Analysis.

Background: Parkinson's disease (PD) has transitioned from a rare condition in 1817 to the fastest-growing neurological disorder globally. The significant increase in cases from 2.5 million in 1990 to 6.1 million in 2016, coupled with predictions of a further doubling by 2040, underscores an impending healthcare challenge. This escalation aligns with global demographic shifts, including rising life expectancy and a growing global population. The economic impact, notably in the U.S., reached $51.9 billion in 2017, with projections suggesting a 46% increase by 2037, emphasizing the substantial socio-economic implications for both patients and caregivers. Coupled with a worldwide demand for health workers that is expected to rise to 80 million by 2030, we have fertile ground for a pandemic. Methods: Our transdisciplinary research focused on early PD detection through running speech and continuous handwriting analysis, incorporating medical, biomedical engineering, AI, and linguistic expertise. The cohort comprised 30 participants, including 20 PD patients at stages 1-4 on the Hoehn and Yahr scale and 10 healthy controls. We employed advanced AI techniques to analyze correlation plots generated from speech and handwriting features, aiming to identify prodromal PD biomarkers. Results: The study revealed distinct speech and handwriting patterns in PD patients compared to controls. Our ParkinsonNet model demonstrated high predictive accuracy, with F1 scores of 95.74% for speech and 96.72% for handwriting analyses. These findings highlight the potential of speech and handwriting as effective early biomarkers for PD. Conclusions: The integration of AI as a decision support system in analyzing speech and handwriting presents a promising approach for early PD detection. This methodology not only offers a novel diagnostic tool but also contributes to the broader understanding of PD's early manifestations. Further research is required to validate these findings in larger, diverse cohorts and to integrate these tools into clinical practice for timely PD pre-diagnosis and management.

Privacy-preserving large language models for structured medical information retrieval

Most clinical information is encoded as free text, not accessible for quantitative analysis. This study presents an open-source pipeline using the local large language model (LLM) “Llama 2” to extract quantitative information from clinical text and evaluates its performance in identifying features of decompensated liver cirrhosis. The LLM identified five key clinical features in a zero- and one-shot manner from 500 patient medical histories in the MIMIC IV dataset. We compared LLMs of three sizes and various prompt engineering approaches, with predictions compared against ground truth from three blinded medical experts. Our pipeline achieved high accuracy, detecting liver cirrhosis with 100% sensitivity and 96% specificity. High sensitivities and specificities were also yielded for detecting ascites (95%, 95%), confusion (76%, 94%), abdominal pain (84%, 97%), and shortness of breath (87%, 97%) using the 70 billion parameter model, which outperformed smaller versions. Our study successfully demonstrates the capability of locally deployed LLMs to extract clinical information from free text with low hardware requirements.

Semi-supervised recognition for artificial intelligence assisted pathology image diagnosis

The analysis and interpretation of cytopathological images are crucial in modern medical diagnostics. However, manually locating and identifying relevant cells from the vast amount of image data can be a daunting task. This challenge is particularly pronounced in developing countries where there may be a shortage of medical expertise to handle such tasks. The challenge of acquiring large amounts of high-quality labelled data remains, many researchers have begun to use semi-supervised learning methods to learn from unlabeled data. Although current semi-supervised learning models partially solve the issue of limited labelled data, they are inefficient in exploiting unlabeled samples. To address this, we introduce a new AI-assisted semi-supervised scheme, the Reliable-Unlabeled Semi-Supervised Segmentation (RU3S) model. This model integrates the ResUNet-SE-ASPP-Attention (RSAA) model, which includes the Squeeze-and-Excitation (SE) network, Atrous Spatial Pyramid Pooling (ASPP) structure, Attention module, and ResUNet architecture. Our model leverages unlabeled data effectively, improving accuracy significantly. A novel confidence filtering strategy is introduced to make better use of unlabeled samples, addressing the scarcity of labelled data. Experimental results show a 2.0% improvement in mIoU accuracy over the current state-of-the-art semi-supervised segmentation model ST, demonstrating our approach’s effectiveness in solving this medical problem.

The use of medicine retailers by people of Goma as an alternative healthcare provider: a risky but rational practice

BackgroundMedicine retailers, considered here as any person or setting dedicated to the sale of retail medicines, fill an important gap in terms of access to healthcare in areas where population are not covered by universal healthcare schemes. In Goma in the Democratic Republic of the Congo, such retailers have proliferated and are consulted as the first port of call by more than half of the population, playing therefore a key role as an alternative source of healthcare for any type of health condition. The objective of this study is to understand people of Goma’s rationale for using the medicine retailers over the formal healthcare system.MethodsTwelve focus groups, gathering 147 participants in total, were conducted in four worship communities, covering the most common faiths practised in Goma. Three focus group discussions were organised per worship community: one with fathers, another with mothers, and another with chronic patients and/or highly vulnerable people. We used a qualitative and inductive approach to analyse the participants’ practices and perceptions in terms of their use of medicine retailers. We identified central categories explaining the reasons for using medicine retailers and the choice of a specific medicine retailer.ResultsWhen facing a health problem, most of the participants in our study tended to first buy medicines at medicine retailers because it was cheap, quick, and easily accessible. Most were aware of the risks and limitations of such practices and had developed a number of mitigation strategies in order to reduce those risks: evaluating medicine retailers’ expertise; developing a “medical expertise”; and seeking proactively out empathetic care.ConclusionsPeople in Goma make a conscious and rational choice when resorting to medicine retailers as it is seen as the least-worst option in a complex situation. In order to reduce the risks, they have developed a number of mitigation strategies. Future research should focus on the organisation of medicine retailers as a professional group to improve their supervision in a sensitive context such as Goma and on modalities to articulate them to the formal health system to guarantee a financial accessibility to healthcare for all.

Assessing The Thermal Damage Induced By Radiofrequency Ablation for Localized Liver Cancer

Abstract Clinical trials are already established for high-temperature treatment of localized cancer, i.e., rise of tissue temperature to more than 550 C as an effective non-invasive method for the treatment of localized cancer. However, as the computational techniques and capacity have enhanced considerably personalized treatment planning has become a manured tool. In the present investigation, a novel treatment planning framework is being proposed for radiofrequency ablation of cancer tissue based on the tomographic image-based actual model. In patient-specific modelling, different thermal parameters like temperature history during ablation, and thermal damage profile have been virtually determined based on Penne's bioheat transfer model with appropriate boundary conditions. This advancement promises to significantly enhance the capabilities of healthcare practitioners in tailoring personalized treatment strategies for their clinical cases. By leveraging simulation outcomes, clinicians can precisely determine the most effective parameters, such as ablation power and frequency. Unfortunately, the current landscape in India presents a scarcity of specialized medical experts in the field of ablation oncology. Moreover, those who practice in this niche often rely on empirical charts rather than data-driven approaches, highlighting a critical need for increased expertise and the integration of advanced simulation technology to optimize cancer tissue ablation procedures. Last but not least optimizing different operating parameters of RF in this patient-centric approach for accurate treatment has also been discussed.

BioPathNet: Enhancing Link Prediction in Biomedical Knowledge Graphs through Path Representation Learning.

Understanding complex interactions in biomedical networks is crucial for advancements in biomedicine, but traditional link prediction (LP) methods are limited in capturing this complexity. Representation-based learning techniques improve prediction accuracy by mapping nodes to low-dimensional embeddings, yet they often struggle with interpretability and scalability. We present BioPathNet, a novel graph neural network framework based on the Neural Bellman-Ford Network (NBFNet), addressing these limitations through path-based reasoning for LP in biomedical knowledge graphs. Unlike node-embedding frameworks, BioPathNet learns representations between node pairs by considering all relations along paths, enhancing prediction accuracy and interpretability. This allows visualization of influential paths and facilitates biological validation. BioPathNet leverages a background regulatory graph (BRG) for enhanced message passing and uses stringent negative sampling to improve precision. In evaluations across various LP tasks, such as gene function annotation, drug-disease indication, synthetic lethality, and lncRNA-mRNA interaction prediction, BioPathNet consistently outperformed shallow node embedding methods, relational graph neural networks and task-specific state-of-the-art methods, demonstrating robust performance and versatility. Our study predicts novel drug indications for diseases like acute lymphoblastic leukemia (ALL) and Alzheimer's, validated by medical experts and clinical trials. We also identified new synthetic lethality gene pairs and regulatory interactions involving lncRNAs and target genes, confirmed through literature reviews. BioPathNet's interpretability will enable researchers to trace prediction paths and gain molecular insights, making it a valuable tool for drug discovery, personalized medicine and biology in general.

Ensemble machine learning framework for predicting maternal health risk during pregnancy

Maternal health risks can cause a range of complications for women during pregnancy. High blood pressure, abnormal glucose levels, depression, anxiety, and other maternal health conditions can all lead to pregnancy complications. Proper identification and monitoring of risk factors can assist to reduce pregnancy complications. The primary goal of this research is to use real-world datasets to identify and predict Maternal Health Risk (MHR) factors. As a result, we developed and implemented the Quad-Ensemble Machine Learning framework to predict Maternal Health Risk Classification (QEML-MHRC). The methodology used a vacxsriety of Machine Learning (ML) models, which then integrated with four ensemble ML techniques to improve prediction. The dataset collected from various maternity hospitals and clinics subjected to nineteen training and testing tests. According to the exploratory data analysis, the most significant risk factors for pregnant women include high blood pressure, low blood pressure, and high blood sugar levels. The study proposed a novel approach to dealing with high-risk factors linked to maternal health. Dealing with class-specific performance elaborated further to properly understand the distinction between high, low, and medium risks. All tests yielded outstanding results when predicting the amount of risk during pregnancy. In terms of class performance, the dataset associated with the “HR” class outperformed the others, predicting 90% correctly. GBT with ensemble stacking outperformed and demonstrated remarkable performance for all evaluation measure (0.86) across all classes in the dataset. The key success of the models used in this work is the ability to measure model performance using a class-wise distribution. The proposed approach can help medical experts assess maternal health risks, saving lives and preventing complications throughout pregnancy. The prediction approach presented in this study can detect high-risk pregnancies early on, allowing for timely intervention and treatment. This study’s development and findings have the potential to raise public awareness of maternal health issues.

Main challenges on the curation of large scale datasets for pancreas segmentation using deep learning in multi-phase CT scans: Focus on cardinality, manual refinement, and annotation quality

Accurate segmentation of the pancreas in computed tomography (CT) holds paramount importance in diagnostics, surgical planning, and interventions. Recent studies have proposed supervised deep-learning models for segmentation, but their efficacy relies on the quality and quantity of the training data. Most of such works employed small-scale public datasets, without proving the efficacy of generalization to external datasets. This study explored the optimization of pancreas segmentation accuracy by pinpointing the ideal dataset size, understanding resource implications, examining manual refinement impact, and assessing the influence of anatomical subregions. We present the AIMS-1300 dataset encompassing 1,300 CT scans. Its manual annotation by medical experts required 938 h. A 2.5D UNet was implemented to assess the impact of training sample size on segmentation accuracy by partitioning the original AIMS-1300 dataset into 11 smaller subsets of progressively increasing numerosity. The findings revealed that training sets exceeding 440 CTs did not lead to better segmentation performance. In contrast, nnU-Net and UNet with Attention Gate reached a plateau for 585 CTs. Tests on generalization on the publicly available AMOS-CT dataset confirmed this outcome. As the size of the partition of the AIMS-1300 training set increases, the number of error slices decreases, reaching a minimum with 730 and 440 CTs, for AIMS-1300 and AMOS-CT datasets, respectively. Segmentation metrics on the AIMS-1300 and AMOS-CT datasets improved more on the head than the body and tail of the pancreas as the dataset size increased. By carefully considering the task and the characteristics of the available data, researchers can develop deep learning models without sacrificing performance even with limited data. This could accelerate developing and deploying artificial intelligence tools for pancreas surgery and other surgical data science applications.

Understanding disciplinary perspectives: a framework to develop skills for interdisciplinary research collaborations of medical experts and engineers

BackgroundHealth professionals need to be prepared for interdisciplinary research collaborations aimed at the development and implementation of medical technology. Expertise is highly domain-specific, and learned by being immersed in professional practice. Therefore, the approaches and results from one domain are not easily understood by experts from another domain. Interdisciplinary collaboration in medical research faces not only institutional, but also cognitive and epistemological barriers. This is one of the reasons why interdisciplinary and interprofessional research collaborations are so difficult. To explain the cognitive and epistemological barriers, we introduce the concept of disciplinary perspectives. Making explicit the disciplinary perspectives of experts participating in interdisciplinary collaborations helps to clarify the specific approach of each expert, thereby improving mutual understanding.MethodWe developed a framework for making disciplinary perspectives of experts participating in an interdisciplinary research collaboration explicit. The applicability of the framework has been tested in an interdisciplinary medical research project aimed at the development and implementation of diffusion MRI for the diagnosis of kidney cancer, where the framework was applied to analyse and articulate the disciplinary perspectives of the experts involved.ResultsWe propose a general framework, in the form of a series of questions, based on new insights from the philosophy of science into the epistemology of interdisciplinary research. We explain these philosophical underpinnings in order to clarify the cognitive and epistemological barriers of interdisciplinary research collaborations. In addition, we present a detailed example of the use of the framework in a concrete interdisciplinary research project aimed at developing a diagnostic technology. This case study demonstrates the applicability of the framework in interdisciplinary research projects.ConclusionInterdisciplinary research collaborations can be facilitated by a better understanding of how an expert’s disciplinary perspectives enables and guides their specific approach to a problem. Implicit disciplinary perspectives can and should be made explicit in a systematic manner, for which we propose a framework that can be used by disciplinary experts participating in interdisciplinary research project. Furthermore, we suggest that educators can explore how the framework and philosophical underpinning can be implemented in HPE to support the development of students’ interdisciplinary expertise.

Weighted graph convolutional network with feature mask for low back pain prediction

Low back pain (LBP) is one of the most common physical disabilities worldwide, imposing a heavy medical economic burden. To help medical experts make disease predictions at an early stage, this paper proposes an LBP abnormality diagnosis model based on a weighted graph convolutional network with feature mask. Unlike popular neural network architectures, we represent the biomechanical characteristics of the spine as a weighted feature graph. The potential relationship between patients is mined through edge connections, and the importance of features for different patients is dynamically captured through the internal feature weighting of nodes. Finally, the importance score of each feature is calculated according to its identification and independence to generate the feature importance mask. It realizes feature sparsification and reinforces important features at the same time, so as to improve the accuracy and stability of the model. The proposed model is verified on a public dataset available in the Kaggle repository, and the classification accuracy achieves 90.2%, which is 2.14% higher than the current best result. Furthermore, when compared for stability against several commonly used machine learning algorithms, the standard deviation of accuracy over ten experiments is only about 1/3 of that of other algorithms.