BreastScreening-AI: Evaluating medical intelligent agents for human-AI interactions

Importance of Patient History in Artificial Intelligence-Assisted Medical Diagnosis: Comparison Study

Background

Medical history contributes approximately 80% to a diagnosis, although physical examinations and laboratory investigations increase a physician's confidence in the medical diagnosis. The concept of artificial intelligence (AI) was first proposed more than 70 years ago. Recently, its role in various fields of medicine has grown remarkably. However, no studies have evaluated the importance of patient history in AI-assisted medical diagnosis.

Objective

This study explored the contribution of patient history to AI-assisted medical diagnoses and assessed the accuracy of ChatGPT in reaching a clinical diagnosis based on the medical history provided.

Methods

Using clinical vignettes of 30 cases identified in The BMJ, we evaluated the accuracy of diagnoses generated by ChatGPT. We compared the diagnoses made by ChatGPT based solely on medical history with the correct diagnoses. We also compared the diagnoses made by ChatGPT after incorporating additional physical examination findings and laboratory data alongside history with the correct diagnoses.

Results

ChatGPT accurately diagnosed 76.6% (23/30) of the cases with only the medical history, consistent with previous research targeting physicians. We also found that this rate was 93.3% (28/30) when additional information was included.

Conclusions

Although adding additional information improves diagnostic accuracy, patient history remains a significant factor in AI-assisted medical diagnosis. Thus, when using AI in medical diagnosis, it is crucial to include pertinent and correct patient histories for an accurate diagnosis. Our findings emphasize the continued significance of patient history in clinical diagnoses in this age and highlight the need for its integration into AI-assisted medical diagnosis systems.

Designing explainable AI to improve human-AI team performance: A medical stakeholder-driven scoping review

The rise of complex AI systems in healthcare and other sectors has led to a growing area of research called Explainable AI (XAI) designed to increase transparency. In this area, quantitative and qualitative studies focus on improving user trust and task performance by providing system- and prediction-level XAI features. We analyze stakeholder engagement events (interviews and workshops) on the use of AI for kidney transplantation. From this we identify themes which we use to frame a scoping literature review on current XAI features. The stakeholder engagement process lasted over nine months covering three stakeholder group's workflows, determining where AI could intervene and assessing a mock XAI decision support system. Based on the stakeholder engagement, we identify four major themes relevant to designing XAI systems - 1) use of AI predictions, 2) information included in AI predictions, 3) personalization of AI predictions for individual differences, and 4) customizing AI predictions for specific cases. Using these themes, our scoping literature review finds that providing AI predictions before, during, or after decision-making could be beneficial depending on the complexity of the stakeholder's task. Additionally, expert stakeholders like surgeons prefer minimal to no XAI features, AI prediction, and uncertainty estimates for easy use cases. However, almost all stakeholders prefer to have optional XAI features to review when needed, especially in hard-to-predict cases. The literature also suggests that providing both system- and prediction-level information is necessary to build the user's mental model of the system appropriately. Although XAI features improve users' trust in the system, human-AI team performance is not always enhanced. Overall, stakeholders prefer to have agency over the XAI interface to control the level of information based on their needs and task complexity. We conclude with suggestions for future research, especially on customizing XAI features based on preferences and tasks.

Understanding the factors influencing acceptability of AI in medical imaging domains among healthcare professionals: A scoping review

BACKGROUND

Artificial intelligence (AI) technology has the potential to transform medical practice within the medical imaging industry and materially improve productivity and patient outcomes. However, low acceptability of AI as a digital healthcare intervention among medical professionals threatens to undermine user uptake levels, hinder meaningful and optimal value-added engagement, and ultimately prevent these promising benefits from being realised. Understanding the factors underpinning AI acceptability will be vital for medical institutions to pinpoint areas of deficiency and improvement within their AI implementation strategies. This scoping review aims to survey the literature to provide a comprehensive summary of the key factors influencing AI acceptability among healthcare professionals in medical imaging domains and the different approaches which have been taken to investigate them.

METHODS

A systematic literature search was performed across five academic databases including Medline, Cochrane Library, Web of Science, Compendex, and Scopus from January 2013 to September 2023. This was done in adherence to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines. Overall, 31 articles were deemed appropriate for inclusion in the scoping review.

RESULTS

The literature has converged towards three overarching categories of factors underpinning AI acceptability including: user factors involving trust, system understanding, AI literacy, and technology receptiveness; system usage factors entailing value proposition, self-efficacy, burden, and workflow integration; and socio-organisational-cultural factors encompassing social influence, organisational readiness, ethicality, and perceived threat to professional identity. Yet, numerous studies have overlooked a meaningful subset of these factors that are integral to the use of medical AI systems such as the impact on clinical workflow practices, trust based on perceived risk and safety, and compatibility with the norms of medical professions. This is attributable to reliance on theoretical frameworks or ad-hoc approaches which do not explicitly account for healthcare-specific factors, the novelties of AI as software as a medical device (SaMD), and the nuances of human-AI interaction from the perspective of medical professionals rather than lay consumer or business end users.

CONCLUSION

This is the first scoping review to survey the health informatics literature around the key factors influencing the acceptability of AI as a digital healthcare intervention in medical imaging contexts. The factors identified in this review suggest that existing theoretical frameworks used to study AI acceptability need to be modified to better capture the nuances of AI deployment in healthcare contexts where the user is a healthcare professional influenced by expert knowledge and disciplinary norms. Increasing AI acceptability among medical professionals will critically require designing human-centred AI systems which go beyond high algorithmic performance to consider accessibility to users with varying degrees of AI literacy, clinical workflow practices, the institutional and deployment context, and the cultural, ethical, and safety norms of healthcare professions. As investment into AI for healthcare increases, it would be valuable to conduct a systematic review and meta-analysis of the causal contribution of these factors to achieving high levels of AI acceptability among medical professionals.

Modeling the influence of attitudes, trust, and beliefs on endoscopists' acceptance of artificial intelligence applications in medical practice

Introduction

The potential for deployment of Artificial Intelligence (AI) technologies in various fields of medicine is vast, yet acceptance of AI amongst clinicians has been patchy. This research therefore examines the role of antecedents, namely trust, attitude, and beliefs in driving AI acceptance in clinical practice.

Methods

We utilized online surveys to gather data from clinicians in the field of gastroenterology.

Results

A total of 164 participants responded to the survey. Participants had a mean age of 44.49 (SD = 9.65). Most participants were male (n = 116, 70.30%) and specialized in gastroenterology (n = 153, 92.73%). Based on the results collected, we proposed and tested a model of AI acceptance in medical practice. Our findings showed that while the proposed drivers had a positive impact on AI tools' acceptance, not all effects were direct. Trust and belief were found to fully mediate the effects of attitude on AI acceptance by clinicians.

Discussion

The role of trust and beliefs as primary mediators of the acceptance of AI in medical practice suggest that these should be areas of focus in AI education, engagement and training. This has implications for how AI systems can gain greater clinician acceptance to engender greater trust and adoption amongst public health systems and professional networks which in turn would impact how populations interface with AI. Implications for policy and practice, as well as future research in this nascent field, are discussed.

Cloud-Based Quad Deep Ensemble Framework for the Detection of COVID-19 Omicron and Delta Variants

The mortality rates of patients contracting the Omicron and Delta variants of COVID-19 are very high, and COVID-19 is the worst variant of COVID. Hence, our objective is to detect COVID-19 Omicron and Delta variants from lung CT-scan images. We designed a unique ensemble model that combines the CNN architecture of a deep neural network-Capsule Network (CapsNet)-and pre-trained architectures, i.e., VGG-16, DenseNet-121, and Inception-v3, to produce a reliable and robust model for diagnosing Omicron and Delta variant data. Despite the solo model's remarkable accuracy, it can often be difficult to accept its results. The ensemble model, on the other hand, operates according to the scientific tenet of combining the majority votes of various models. The adoption of the transfer learning model in our work is to benefit from previously learned parameters and lower data-hunger architecture. Likewise, CapsNet performs consistently regardless of positional changes, size changes, and changes in the orientation of the input image. The proposed ensemble model produced an accuracy of 99.93%, an AUC of 0.999 and a precision of 99.9%. Finally, the framework is deployed in a local cloud web application so that the diagnosis of these particular variants can be accomplished remotely.

The Iterative Design Process of an Explainable AI Application for Non-Invasive Diagnosis of CNS Tumors: A User-Centered Approach

Artificial Intelligence (AI) is well-suited to help support complex decision-making tasks within clinical medicine, including clinical imaging applications like radiographic differential diagnosis of central nervous system (CNS) tumors. So far, there have been numerous examples of theoretical AI solutions for this space, for example, large-scale corporate efforts like IBM's Watson AI. However, clinical implementation remains limited due to factors related to the alignment of this technology in the clinical setting. User-Centered Design (UCD) is a design philosophy that focuses on developing tailored solutions for specific users or user groups. In this study, we applied UCD to develop an explainable AI tool to support clinicians in our use case. Through four design iterations, starting from basic functionality and visualizations, we progressed to functional prototypes in a realistic testing environment. We discuss our motivation and approach for each iteration, along with key insights gained. This UCD process has advanced our conceptual idea from feasibility testing to interactive functional AI interfaces designed for specific clinical and cognitive tasks. It has also provided us with directions to develop further an AI system for the non-invasive diagnosis of CNS tumors.

Plant disease prescription recommendation based on electronic medical records and sentence embedding retrieval

BACKGROUND

In the era of Agri 4.0 and the popularity of Plantwise systems, the availability of Plant Electronic Medical Records has provided opportunities to extract valuable disease information and treatment knowledge. However, developing an effective prescription recommendation method based on these records presents unique challenges, such as inadequate labeling data, lack of structural and linguistic specifications, incorporation of new prescriptions, and consideration of multiple factors in practical situations.

RESULTS

This study proposes a plant disease prescription recommendation method called PRSER, which is based on sentence embedding retrieval. The semantic matching model is created using a pre-trained language model and a sentence embedding method with contrast learning ideas, and the constructed prescription reference database is retrieved for optimal prescription recommendations. A multi-vegetable disease dataset and a multi-fruit disease dataset are constructed to compare three pre-trained language models, four pooling types, and two loss functions. The PRSER model achieves the best semantic matching performance by combining MacBERT, CoSENT, and CLS pooling, resulting in a Pearson coefficient of 86.34% and a Spearman coefficient of 77.67%. The prescription recommendation capability of the model is also verified. PRSER performs well in closed-set testing with Top-1/Top-3/Top-5 accuracy of 88.20%/96.07%/97.70%; and slightly worse in open-set testing with Top-1/Top-3/Top-5 accuracy of 82.04%/91.50%/94.90%. Finally, a plant disease prescription recommendation system for mobile terminals is constructed and its generalization ability with incomplete inputs is verified. When only symptom information is available without environment and plant information, our model shows slightly lower accuracy with Top-1/Top-3/Top-5 accuracy of 75.24%/88.35%/91.99% in closed-set testing and Top-1/Top-3/Top-5 accuracy of 75.08%/87.54%/89.84% in open-set testing.

CONCLUSIONS

The experiments validate the effectiveness and generalization ability of the proposed approach for recommending plant disease prescriptions. This research has significant potential to facilitate the implementation of artificial intelligence in plant disease treatment, addressing the needs of farmers and advancing scientific plant disease management.

Generating post-hoc explanations for Skip-gram-based node embeddings by identifying important nodes with bridgeness

Node representation learning in a network is an important machine learning technique for encoding relational information in a continuous vector space while preserving the inherent properties and structures of the network. Recently, unsupervised node embedding methods such as DeepWalk (Perozzi et al., 2014), LINE (Tang et al., 2015), struc2vec (Ribeiro et al., 2017), PTE (Tang et al., 2015), UserItem2vec (Wu et al., 2020), and RWJBG (Li et al., 2021) have emerged from the Skip-gram model (Mikolov et al., 2013) and perform better performance in several downstream tasks such as node classification and link prediction than the existing relational models. However, providing post-hoc explanations of unsupervised embeddings remains a challenging problem because of the lack of explanation methods and theoretical studies applicable for embeddings. In this paper, we first show that global explanations to the Skip-gram-based embeddings can be found by computing bridgeness under a spectral cluster-aware local perturbation. Moreover, a novel gradient-based explanation method, which we call GRAPH-wGD, is proposed that allows the top-q global explanations about learned graph embedding vectors more efficiently. Experiments show that the ranking of nodes by scores using GRAPH-wGD is highly correlated with true bridgeness scores. We also observe that the top-q node-level explanations selected by GRAPH-wGD have higher importance scores and produce more changes in class label prediction when perturbed, compared with the nodes selected by recent alternatives, using five real-world graphs.

The Role of AI in Breast Cancer Lymph Node Classification: A Comprehensive Review

Breast cancer is a significant health issue affecting women worldwide, and accurately detecting lymph node metastasis is critical in determining treatment and prognosis. While traditional diagnostic methods have limitations and complications, artificial intelligence (AI) techniques such as machine learning (ML) and deep learning (DL) offer promising solutions for improving and supplementing diagnostic procedures. Current research has explored state-of-the-art DL models for breast cancer lymph node classification from radiological images, achieving high performances (AUC: 0.71-0.99). AI models trained on clinicopathological features also show promise in predicting metastasis status (AUC: 0.74-0.77), whereas multimodal (radiomics + clinicopathological features) models combine the best from both approaches and also achieve good results (AUC: 0.82-0.94). Once properly validated, such models could greatly improve cancer care, especially in areas with limited medical resources. This comprehensive review aims to compile knowledge about state-of-the-art AI models used for breast cancer lymph node metastasis detection, discusses proper validation techniques and potential pitfalls and limitations, and presents future directions and best practices to achieve high usability in real-world clinical settings.

Improving Automatic Melanoma Diagnosis Using Deep Learning-Based Segmentation of Irregular Networks

Deep learning has achieved significant success in malignant melanoma diagnosis. These diagnostic models are undergoing a transition into clinical use. However, with melanoma diagnostic accuracy in the range of ninety percent, a significant minority of melanomas are missed by deep learning. Many of the melanomas missed have irregular pigment networks visible using dermoscopy. This research presents an annotated irregular network database and develops a classification pipeline that fuses deep learning image-level results with conventional hand-crafted features from irregular pigment networks. We identified and annotated 487 unique dermoscopic melanoma lesions from images in the ISIC 2019 dermoscopic dataset to create a ground-truth irregular pigment network dataset. We trained multiple transfer learned segmentation models to detect irregular networks in this training set. A separate, mutually exclusive subset of the International Skin Imaging Collaboration (ISIC) 2019 dataset with 500 melanomas and 500 benign lesions was used for training and testing deep learning models for the binary classification of melanoma versus benign. The best segmentation model, U-Net++, generated irregular network masks on the 1000-image dataset. Other classical color, texture, and shape features were calculated for the irregular network areas. We achieved an increase in the recall of melanoma versus benign of 11% and in accuracy of 2% over DL-only models using conventional classifiers in a sequential pipeline based on the cascade generalization framework, with the highest increase in recall accompanying the use of the random forest algorithm. The proposed approach facilitates leveraging the strengths of both deep learning and conventional image processing techniques to improve the accuracy of melanoma diagnosis. Further research combining deep learning with conventional image processing on automatically detected dermoscopic features is warranted.

Detecting mild cognitive impairment and dementia in older adults using naturalistic driving data and interaction-based classification from influence score

Several recent studies indicate that atypical changes in driving behaviors appear to be early signs of mild cognitive impairment (MCI) and dementia. These studies, however, are limited by small sample sizes and short follow-up duration. This study aims to develop an interaction-based classification method building on a statistic named Influence Score (i.e., I-score) for prediction of MCI and dementia using naturalistic driving data collected from the Longitudinal Research on Aging Drivers (LongROAD) project. Naturalistic driving trajectories were collected through in-vehicle recording devices for up to 44 months from 2977 participants who were cognitively intact at the time of enrollment. These data were further processed and aggregated to generate 31 time-series driving variables. Because of high dimensional time-series features for driving variables, we used I-score for variable selection. I-score is a measure to evaluate variables' ability to predict and is proven to be effective in differentiating between noisy and predictive variables in big data. It is introduced here to select influential variable modules or groups that account for compound interactions among explanatory variables. It is explainable regarding to what extent variables and their interactions contribute to the predictiveness of a classifier. In addition, I-score boosts the performance of classifiers over imbalanced datasets due to its association with the F1 score. Using predictive variables selected by I-score, interaction-based residual blocks are constructed over top I-score modules to generate predictors and ensemble learning aggregates these predictors to boost the prediction of the overall classifier. Experiments using naturalistic driving data show that our proposed classification method achieves the best accuracy (96%) for predicting MCI and dementia, followed by random forest (93%) and logistic regression (88%). In terms of F1 score and AUC, our proposed classifier achieves 98% and 87%, respectively, followed by random forest (with an F1 score of 96% and an AUC of 79%) and logistic regression (with an F1 score of 92% and an AUC of 77%). The results indicate that incorporating I-score into machine learning algorithms could considerably improve the model performance for predicting MCI and dementia in older drivers. We also performed the feature importance analysis and found that the right to left turn ratio and the number of hard braking events are the most important driving variables to predict MCI and dementia.

Pancreatic cancer grading in pathological images using deep learning convolutional neural networks

Background: Pancreatic cancer is one of the deadliest forms of cancer. The cancer grades define how aggressively the cancer will spread and give indication for doctors to make proper prognosis and treatment. The current method of pancreatic cancer grading, by means of manual examination of the cancerous tissue following a biopsy, is time consuming and often results in misdiagnosis and thus incorrect treatment. This paper presents an automated grading system for pancreatic cancer from pathology images developed by comparing deep learning models on two different pathological stains. Methods: A transfer-learning technique was adopted by testing the method on 14 different ImageNet pre-trained models. The models were fine-tuned to be trained with our dataset. Results: From the experiment, DenseNet models appeared to be the best at classifying the validation set with up to 95.61% accuracy in grading pancreatic cancer despite the small sample set. Conclusions: To the best of our knowledge, this is the first work in grading pancreatic cancer based on pathology images. Previous works have either focused only on detection (benign or malignant), or on radiology images (computerized tomography [CT], magnetic resonance imaging [MRI] etc.). The proposed system can be very useful to pathologists in facilitating an automated or semi-automated cancer grading system, which can address the problems found in manual grading.

Systematic analysis of the test design and performance of AI/ML-based medical devices approved for triage/detection/diagnosis in the USA and Japan

The development of computer-aided detection (CAD) using artificial intelligence (AI) and machine learning (ML) is rapidly evolving. Submission of AI/ML-based CAD devices for regulatory approval requires information about clinical trial design and performance criteria, but the requirements vary between countries. This study compares the requirements for AI/ML-based CAD devices approved by the US Food and Drug Administration (FDA) and the Pharmaceuticals and Medical Devices Agency (PMDA) in Japan. A list of 45 FDA-approved and 12 PMDA-approved AI/ML-based CAD devices was compiled. In the USA, devices classified as computer-aided simple triage were approved based on standalone software testing, whereas devices classified as computer-aided detection/diagnosis were approved based on reader study testing. In Japan, however, there was no clear distinction between evaluation methods according to the category. In the USA, a prospective randomized controlled trial was conducted for AI/ML-based CAD devices used for the detection of colorectal polyps, whereas in Japan, such devices were approved based on standalone software testing. This study indicated that the different viewpoints of AI/ML-based CAD in the two countries influenced the selection of different evaluation methods. This study’s findings may be useful for defining a unified global development and approval standard for AI/ML-based CAD.

Automated identification of hip arthroplasty implants using artificial intelligence

The purpose of this study was to develop and evaluate the performance of deep learning methods based on convolutional neural networks (CNN) to detect and identify specific hip arthroplasty models. In this study, we propose a novel deep learning-based approach to identify hip arthroplasty implants' design using anterior-posterior images of both the stem and the cup. We harness the pre-trained ResNet50 CNN model and employ transfer learning methods to adapt the model for the implants identification task using a total of 714 radiographs of 4 different hip arthroplasty implant designs. Performance was compared with the operative notes and crosschecked with implant sheets. We also evaluate the difference in performance of models trained with the images of the stem, the cup or both. The training and validation data sets were comprised of 357 stem images and 357 cup radiographs across 313 patients and included 4 hip arthroplasty implants from 4 leading implant manufacturers. After 1000 training epochs the model classified 4 implant models with very high accuracy. Our results showed that jointly using stem images and cup images did not improve the classification accuracy of the CNN model. CNN can accurately distinguish between specific hip arthroplasty designs. This technology could offer a useful adjunct to the surgeon in preoperative identification of the prior implant. Using stem images or cup images to train the CNN can both achieve effective identification accuracy, with the accuracy of the stem images being higher. Using stem images and cup images together is not more effective than using images from only one perspective.