Artificial Intelligence Applications in Dermatology: Where Do We Stand?

Design and development of a machine-learning-driven opioid overdose risk prediction tool integrated in electronic health records in primary care settings

BACKGROUND

Integrating advanced machine-learning (ML) algorithms into clinical practice is challenging and requires interdisciplinary collaboration to develop transparent, interpretable, and ethically sound clinical decision support (CDS) tools. We aimed to design a ML-driven CDS tool to predict opioid overdose risk and gather feedback for its integration into the University of Florida Health (UFHealth) electronic health record (EHR) system.

METHODS

We used user-centered design methods to integrate the ML algorithm into the EHR system. The backend and UI design sub-teams collaborated closely, both informed by user feedback sessions. We conducted seven user feedback sessions with five UF Health primary care physicians (PCPs) to explore aspects of CDS tools, including workflow, risk display, and risk mitigation strategies. After customizing the tool based on PCPs' feedback, we held two rounds of one-on-one usability testing sessions with 8 additional PCPs to gather feedback on prototype alerts. These sessions informed iterative UI design and backend processes, including alert frequency and reappearance circumstances.

RESULTS

The backend process development identified needs and requirements from our team, information technology, UFHealth, and PCPs. Thirteen PCPs (male = 62%, White = 85%) participated across 7 user feedback sessions and 8 usability testing sessions. During the user feedback sessions, PCPs (n = 5) identified flaws such as the term "high risk" of overdose potentially leading to unintended consequences (e.g., immediate addiction services referrals), offered suggestions, and expressed trust in the tool. In the first usability testing session, PCPs (n = 4) emphasized the need for natural risk presentation (e.g., 1 in 200) and suggested displaying the alert multiple times yearly for at-risk patients. Another 4 PCPs in the second usability testing session valued the UFHealth-specific alert for managing new or unfamiliar patients, expressed concerns about PCPs' workload when prescribing to high-risk patients, and recommended incorporating the details page into training sessions to enhance usability.

CONCLUSIONS

The final backend process for our CDS alert aligns with PCP needs and UFHealth standards. Integrating feedback from PCPs in the early development phase of our ML-driven CDS tool helped identify barriers and facilitators in the CDS integration process. This collaborative approach yielded a refined prototype aimed at minimizing unintended consequences and enhancing usability.

AI in Psoriatic Disease: Scoping Review

BACKGROUND

Artificial intelligence (AI) has many applications in numerous medical fields, including dermatology. Although the majority of AI studies in dermatology focus on skin cancer, there is growing interest in the applicability of AI models in inflammatory diseases, such as psoriasis. Psoriatic disease is a chronic, inflammatory, immune-mediated systemic condition with multiple comorbidities and a significant impact on patients' quality of life. Advanced treatments, including biologics and small molecules, have transformed the management of psoriatic disease. Nevertheless, there are still considerable unmet needs. Globally, delays in the diagnosis of the disease and its severity are common due to poor access to health care systems. Moreover, despite the abundance of treatments, we are unable to predict which is the right medication for the right patient, especially in resource-limited settings. AI could be an additional tool to address those needs. In this way, we can improve rates of diagnosis, accurately assess severity, and predict outcomes of treatment.

OBJECTIVE

This study aims to provide an up-to-date literature review on the use of AI in psoriatic disease, including diagnostics and clinical management as well as addressing the limitations in applicability.

METHODS

We searched the databases MEDLINE, PubMed, and Embase using the keywords "AI AND psoriasis OR psoriatic arthritis OR psoriatic disease," "machine learning AND psoriasis OR psoriatic arthritis OR psoriatic disease," and "prognostic model AND psoriasis OR psoriatic arthritis OR psoriatic disease" until June 1, 2023. Reference lists of relevant papers were also cross-examined for other papers not detected in the initial search.

RESULTS

Our literature search yielded 38 relevant papers. AI has been identified as a key component in digital health technologies. Within this field, there is the potential to apply specific techniques such as machine learning and deep learning to address several aspects of managing psoriatic disease. This includes diagnosis, particularly useful for remote teledermatology via photographs taken by patients as well as monitoring and estimating severity. Similarly, AI can be used to synthesize the vast data sets already in place through patient registries which can help identify appropriate biologic treatments for future cohorts and those individuals most likely to develop complications.

CONCLUSIONS

There are multiple advantageous uses for AI and digital health technologies in psoriatic disease. With wider implementation of AI, we need to be mindful of potential limitations, such as validation and standardization or generalizability of results in specific populations, such as patients with darker skin phototypes.

Assessing the Impact of ChatGPT in Dermatology: A Comprehensive Rapid Review

Background/Objectives: The use of artificial intelligence (AI) in dermatology is expanding rapidly, with ChatGPT, a large language model (LLM) from OpenAI, showing promise in patient education, clinical decision-making, and teledermatology. Despite its potential, the ethical, clinical, and practical implications of its application remain insufficiently explored. This study aims to evaluate the effectiveness, challenges, and future prospects of ChatGPT in dermatology, focusing on clinical applications, patient interactions, and medical writing. ChatGPT was selected due to its broad adoption, extensive validation, and strong performance in dermatology-related tasks. Methods: A thorough literature review was conducted, focusing on publications related to ChatGPT and dermatology. The search included articles in English from November 2022 to August 2024, as this period captures the most recent developments following the launch of ChatGPT in November 2022, ensuring that the review includes the latest advancements and discussions on its role in dermatology. Studies were chosen based on their relevance to clinical applications, patient interactions, and ethical issues. Descriptive metrics, such as average accuracy scores and reliability percentages, were used to summarize study characteristics, and key findings were analyzed. Results: ChatGPT has shown significant potential in passing dermatology specialty exams and providing reliable responses to patient queries, especially for common dermatological conditions. However, it faces limitations in diagnosing complex cases like cutaneous neoplasms, and concerns about the accuracy and completeness of its information persist. Ethical issues, including data privacy, algorithmic bias, and the need for transparent guidelines, were identified as critical challenges. Conclusions: While ChatGPT has the potential to significantly enhance dermatological practice, particularly in patient education and teledermatology, its integration must be cautious, addressing ethical concerns and complementing, rather than replacing, dermatologist expertise. Future research should refine ChatGPT's diagnostic capabilities, mitigate biases, and develop comprehensive clinical guidelines.

Artificial intelligence for nonmelanoma skin cancer

Nonmelanoma skin cancers (NMSCs) are among the top five most common cancers globally. NMSC is an area with great potential for novel application of diagnostic tools including artificial intelligence (AI). In this scoping review, we aimed to describe the applications of AI in the diagnosis and treatment of NMSC. Twenty-nine publications described AI applications to dermatopathology including lesion classification and margin assessment. Twenty-five publications discussed AI use in clinical image analysis, showing that algorithms are not superior to dermatologists and may rely on unbalanced, nonrepresentative, and nontransparent training data sets. Sixteen publications described the use of AI in cutaneous surgery for NMSC including use in margin assessment during excisions and Mohs surgery, as well as predicting procedural complexity. Eleven publications discussed spectroscopy, confocal microscopy, thermography, and the AI algorithms that analyze and interpret their data. Ten publications pertained to AI applications for the discovery and use of NMSC biomarkers. Eight publications discussed the use of smartphones and AI, specifically how they enable clinicians and patients to have increased access to instant dermatologic assessments but with varying accuracies. Five publications discussed large language models and NMSC, including how they may facilitate or hinder patient education and medical decision-making. Three publications pertaining to the skin of color and AI for NMSC discussed concerns regarding limited diverse data sets for the training of convolutional neural networks. AI demonstrates tremendous potential to improve diagnosis, patient and clinician education, and management of NMSC. Despite excitement regarding AI, data sets are often not transparently reported, may include low-quality images, and may not include diverse skin types, limiting generalizability. AI may serve as a tool to increase access to dermatology services for patients in rural areas and save health care dollars. These benefits can only be achieved, however, with consideration of potential ethical costs.

Deep learning algorithms for melanoma detection using dermoscopic images: A systematic review and meta-analysis

BACKGROUND

Melanoma is a serious risk to human health and early identification is vital for treatment success. Deep learning (DL) has the potential to detect cancer using imaging technologies and many studies provide evidence that DL algorithms can achieve high accuracy in melanoma diagnostics.

OBJECTIVES

To critically assess different DL performances in diagnosing melanoma using dermatoscopic images and discuss the relationship between dermatologists and DL.

METHODS

Ovid-Medline, Embase, IEEE Xplore, and the Cochrane Library were systematically searched from inception until 7th December 2021. Studies that reported diagnostic DL model performances in detecting melanoma using dermatoscopic images were included if they had specific outcomes and histopathologic confirmation. Binary diagnostic accuracy data and contingency tables were extracted to analyze outcomes of interest, which included sensitivity (SEN), specificity (SPE), and area under the curve (AUC). Subgroup analyses were performed according to human-machine comparison and cooperation. The study was registered in PROSPERO, CRD42022367824.

RESULTS

2309 records were initially retrieved, of which 37 studies met our inclusion criteria, and 27 provided sufficient data for meta-analytical synthesis. The pooled SEN was 82 % (range 77-86), SPE was 87 % (range 84-90), with an AUC of 0.92 (range 0.89-0.94). Human-machine comparison had pooled AUCs of 0.87 (0.84-0.90) and 0.83 (0.79-0.86) for DL and dermatologists, respectively. Pooled AUCs were 0.90 (0.87-0.93), 0.80 (0.76-0.83), and 0.88 (0.85-0.91) for DL, and junior and senior dermatologists, respectively. Analyses of human-machine cooperation were 0.88 (0.85-0.91) for DL, 0.76 (0.72-0.79) for unassisted, and 0.87 (0.84-0.90) for DL-assisted dermatologists.

CONCLUSIONS

Evidence suggests that DL algorithms are as accurate as senior dermatologists in melanoma diagnostics. Therefore, DL could be used to support dermatologists in diagnostic decision-making. Although, further high-quality, large-scale multicenter studies are required to address the specific challenges associated with medical AI-based diagnostics.

Artificial intelligence in autoimmune bullous dermatoses

Dermatologists treating patients with autoimmune bullous dermatoses (AIBDs), as well as the patients themselves, encounter challenges at every stage of their interaction, including dermatologic and comorbidities assessment, diagnosis, prognosis evaluation, treatment, and follow-up monitoring. We summarize the current and potential future clinical applications of artificial intelligence (AI) in the field of AIBDs. Recent research and AI models have demonstrated their potential to enhance or may already be contributing to advancements in every phase of the comprehensive diagnosis and personalized treatment process in AIBDs, providing patients, clinicians, and administrators with valuable support. Image recognition AI systems might assist precise clinical diagnoses of various diseases, including AIBDs, and could offer consistent and reliable scoring of disease severity. Automated and standardized AI-assisted laboratory methods could improve the accuracy and decrease the time and cost of gold-standard tests such as direct and indirect immunofluorescence. The studies and tools discussed in this contribution, although in the early stages, might be a small precursor to a transformative shift in the way we take care of patients with chronic skin diseases, including AIBDs.

Claude 3 Opus and ChatGPT With GPT-4 in Dermoscopic Image Analysis for Melanoma Diagnosis: Comparative Performance Analysis

BACKGROUND

Recent advancements in artificial intelligence (AI) and large language models (LLMs) have shown potential in medical fields, including dermatology. With the introduction of image analysis capabilities in LLMs, their application in dermatological diagnostics has garnered significant interest. These capabilities are enabled by the integration of computer vision techniques into the underlying architecture of LLMs.

OBJECTIVE

This study aimed to compare the diagnostic performance of Claude 3 Opus and ChatGPT with GPT-4 in analyzing dermoscopic images for melanoma detection, providing insights into their strengths and limitations.

METHODS

We randomly selected 100 histopathology-confirmed dermoscopic images (50 malignant, 50 benign) from the International Skin Imaging Collaboration (ISIC) archive using a computer-generated randomization process. The ISIC archive was chosen due to its comprehensive and well-annotated collection of dermoscopic images, ensuring a diverse and representative sample. Images were included if they were dermoscopic images of melanocytic lesions with histopathologically confirmed diagnoses. Each model was given the same prompt, instructing it to provide the top 3 differential diagnoses for each image, ranked by likelihood. Primary diagnosis accuracy, accuracy of the top 3 differential diagnoses, and malignancy discrimination ability were assessed. The McNemar test was chosen to compare the diagnostic performance of the 2 models, as it is suitable for analyzing paired nominal data.

RESULTS

In the primary diagnosis, Claude 3 Opus achieved 54.9% sensitivity (95% CI 44.08%-65.37%), 57.14% specificity (95% CI 46.31%-67.46%), and 56% accuracy (95% CI 46.22%-65.42%), while ChatGPT demonstrated 56.86% sensitivity (95% CI 45.99%-67.21%), 38.78% specificity (95% CI 28.77%-49.59%), and 48% accuracy (95% CI 38.37%-57.75%). The McNemar test showed no significant difference between the 2 models (P=.17). For the top 3 differential diagnoses, Claude 3 Opus and ChatGPT included the correct diagnosis in 76% (95% CI 66.33%-83.77%) and 78% (95% CI 68.46%-85.45%) of cases, respectively. The McNemar test showed no significant difference (P=.56). In malignancy discrimination, Claude 3 Opus outperformed ChatGPT with 47.06% sensitivity, 81.63% specificity, and 64% accuracy, compared to 45.1%, 42.86%, and 44%, respectively. The McNemar test showed a significant difference (P<.001). Claude 3 Opus had an odds ratio of 3.951 (95% CI 1.685-9.263) in discriminating malignancy, while ChatGPT-4 had an odds ratio of 0.616 (95% CI 0.297-1.278).

CONCLUSIONS

Our study highlights the potential of LLMs in assisting dermatologists but also reveals their limitations. Both models made errors in diagnosing melanoma and benign lesions. These findings underscore the need for developing robust, transparent, and clinically validated AI models through collaborative efforts between AI researchers, dermatologists, and other health care professionals. While AI can provide valuable insights, it cannot yet replace the expertise of trained clinicians.

Automated Machine Learning (AutoML) for the Diagnosis of Melanoma Skin Lesions From Consumer-Grade Camera Photos

BACKGROUND

In recent years, there has been much speculation about the role of artificial intelligence (AI) and machine learning in dermatology. Advances in computer vision have increased the potential for automated diagnosis of images. However, there remains a gap between the technological development of the algorithms and their real-world implementation. This study aims to develop and test an automated machine learning (AutoML) algorithm for the diagnosis of melanoma, with no technical or coding skills required by the operator.

METHODS

The Skin Cancer Detection Dataset from the University of Waterloo Vision and Image Processing Lab contains 206 images sourced from the public databases DermIS and DermQuest. The dataset was split into two groups: training data (n=174) and testing data (n=32). A machine learning algorithm was created using 'Teachable Machine', trained on the training data, to differentiate between melanoma and non-melanoma skin lesions.

RESULTS

The AutoML algorithm identified 12/14 non-melanoma images and 15/18 melanoma images in the testing dataset. The overall accuracy was 84.4%, with a sensitivity of 83.3% and a specificity of 85.7%.

CONCLUSIONS

Existing literature has tested a range of different machine learning algorithms on the same dataset. These have often required expertise in machine learning and the ability to code. The results of this study, using a no-code tool, perform comparably to existing efforts and suggest that there is potential for future clinical AI algorithms to be developed by doctors even without any technical expertise as long as they have access to relevant local data.

Automated Image Quality and Protocol Adherence Assessment of Examinations in Teledermatology: First Results

Introduction: Image quality and acquisition protocol adherence assessment is a neglected area in teledermatology. We examine if it is feasible to use deep learning methods to automate the assessment of the adherence of examinations to image acquisition protocols. In this study, we focused on the quality criteria of two image acquisition protocols: (1) approximation image and (2) panoramic image, as these are present in all teledermatology examination protocols currently used by the Santa Catarina State Integrated Telemedicine and Telehealth System (STT/SC). Methods: We use a data set of 36,102 teledermatological examinations performed at the STT/SC during 2021. As our validation process, we adopted standard machine learning metrics and an inter-rater agreement (IRA) study with 11 dermatologists. For the approximation image protocol, we used the Mask-Region based Convolutional Neural Network (RCNN) Object Detection Deep Learning (DL) architecture to identify the presence of a lesion identification tag and a ruler used to provide a frame reference of the lesion. For the panoramic image protocol, we used DensePose, a pose estimation DL, architecture to assess the presence of a whole patient body and its orientation. A combination of the two approaches was additionally validated through an IRA study between specialists. Results: Mask-RCNN achieved a score of 96% mean average precision (mAP), while DensePose presented 75% mAP. IRA achieved a level of agreement of 96.68% with the Krippendorff alpha score. Conclusions: Our results show the feasibility of using deep learning to automate the image quality and protocol adherence assessment in teledermatology, before the specialist's manual analysis of the examination.

The State of Artificial Intelligence in Skin Cancer Publications

BACKGROUND

Artificial intelligence (AI) in skin cancer is a promising research field to assist physicians and to provide support to patients remotely. Physicians' awareness to new developments in AI research is important to define the best practices and scope of integrating AI-enabled technologies within a clinical setting.

OBJECTIVES

To analyze the characteristics and trends of AI skin cancer publications from dermatology journals.

METHODS

AI skin cancer publications were retrieved in June 2022 from the Web of Science. Publications were screened by title, abstract, and keywords to assess eligibility. Publications were fully reviewed. Publications were divided between nonmelanoma skin cancer (NMSC), melanoma, and skin cancer studies. The primary measured outcome was the number of citations. The secondary measured outcomes were articles' general characteristics and features related to AI.

RESULTS

A total of 168 articles were included: 25 on NMSC, 77 on melanoma, and 66 on skin cancer. The most common types of skin cancers were melanoma (134, 79.8%), basal cell carcinoma (61, 36.3%), and squamous cell carcinoma (45, 26.9%). All articles were published between 2000 and 2022, with 49 (29.2%) of them being published in 2021. Original studies that developed or assessed an algorithm predominantly used supervised learning (66, 97.0%) and deep neural networks (42, 67.7%). The most used imaging modalities were standard dermoscopy (76, 45.2%) and clinical images (39, 23.2%).

CONCLUSIONS

Most publications focused on developing or assessing screening technologies with mainly deep neural network algorithms. This indicates the eminent need for dermatologists to label or annotate images used by novel AI systems.

Novel Discoveries and Clinical Advancements for Treating Onychomycosis: A Mechanistic Insight

Onychomycosis continues to be the most challenging disease condition for pharmaceutical scientists to develop an effective drug delivery system. Treatment challenges lie in incomplete cure and high relapse rate. Present compilation provides cumulative information on pathophysiology, diagnostic techniques, and conventional treatment strategies to manage onychomycosis. Novel technologies developed for successful delivery of antifungal molecules are also discussed in brief. Multidirectional information offered by this article also unlocks the panoramic view of leading patented technologies and clinical trials. The obtained clinical landscape recommends the use of advanced technology driven approaches, as a promising way-out for treatment of onychomycosis. Collectively, present review warrants the application of novel technologies for the successful management of onychomycosis. This review will assist readers to envision a better understanding about the technologies available for combating onychomycosis. We also trust that these contributions address and certainly will encourage the design and development of nanocarriers-based delivery vehicles for effective management of onychomycosis.

Cutaneous Oncology: Strategies for Melanoma Prevention, Diagnosis, and Therapy

Skin cancer comprises one-third of all diagnosed cancer cases and remains a major health concern. Genetic and environmental parameters serve as the two main risk factors associated with the development of skin cancer, with ultraviolet radiation being the most common environmental risk factor. Studies have also found fair complexion, arsenic toxicity, indoor tanning, and family history among the prevailing causes of skin cancer. Prevention and early diagnosis play a crucial role in reducing the frequency and ensuring effective management of skin cancer. Recent studies have focused on exploring minimally invasive or non-invasive diagnostic technologies along with artificial intelligence to facilitate rapid and accurate diagnosis. The treatment of skin cancer ranges from traditional surgical excision to various advanced methods such as phototherapy, radiotherapy, immunotherapy, targeted therapy, and combination therapy. Recent studies have focused on immunotherapy, with the introduction of new checkpoint inhibitors and personalized immunotherapy enhancing treatment efficacy. Advancements in multi-omics, nanotechnology, and artificial intelligence have further deepened the understanding of the mechanisms underlying tumoral growth and their interaction with therapeutic effects, which has paved the way for precision oncology. This review aims to highlight the recent advancements in the understanding and management of skin cancer, and provide an overview of existing and emerging diagnostic, prognostic, and therapeutic modalities, while highlighting areas that require further research to bridge the existing knowledge gaps.

Principles, applications, and future of artificial intelligence in dermatology

This paper provides an overview of artificial-intelligence (AI), as applied to dermatology. We focus our discussion on methodology, AI applications for various skin diseases, limitations, and future opportunities. We review how the current image-based models are being implemented in dermatology across disease subsets, and highlight the challenges facing widespread adoption. Additionally, we discuss how the future of AI in dermatology might evolve and the emerging paradigm of large language, and multi-modal models to emphasize the importance of developing responsible, fair, and equitable models in dermatology.

A Systematic Review of the Barriers to the Implementation of Artificial Intelligence in Healthcare

Artificial intelligence (AI) is expected to improve healthcare outcomes by facilitating early diagnosis, reducing the medical administrative burden, aiding drug development, personalising medical and oncological management, monitoring healthcare parameters on an individual basis, and allowing clinicians to spend more time with their patients. In the post-pandemic world where there is a drive for efficient delivery of healthcare and manage long waiting times for patients to access care, AI has an important role in supporting clinicians and healthcare systems to streamline the care pathways and provide timely and high-quality care for the patients. Despite AI technologies being used in healthcare for some decades, and all the theoretical potential of AI, the uptake in healthcare has been uneven and slower than anticipated and there remain a number of barriers, both overt and covert, which have limited its incorporation. This literature review highlighted barriers in six key areas: ethical, technological, liability and regulatory, workforce, social, and patient safety barriers. Defining and understanding the barriers preventing the acceptance and implementation of AI in the setting of healthcare will enable clinical staff and healthcare leaders to overcome the identified hurdles and incorporate AI technologies for the benefit of patients and clinical staff.

Exploring the potential of artificial intelligence in improving skin lesion diagnosis in primary care

Dermatological conditions are a relevant health problem. Machine learning (ML) models are increasingly being applied to dermatology as a diagnostic decision support tool using image analysis, especially for skin cancer detection and disease classification. The objective of this study was to perform a prospective validation of an image analysis ML model, which is capable of screening 44 skin diseases, comparing its diagnostic accuracy with that of General Practitioners (GPs) and teledermatology (TD) dermatologists in a real-life setting. Prospective, diagnostic accuracy study including 100 consecutive patients with a skin problem who visited a participating GP in central Catalonia, Spain, between June 2021 and October 2021. The skin issue was first assessed by the GPs. Then an anonymised skin disease picture was taken and uploaded to the ML application, which returned a list with the Top-5 possible diagnosis in order of probability. The same image was then sent to a dermatologist via TD for diagnosis, as per clinical practice. The GPs Top-3, ML model's Top-5 and dermatologist's Top-3 assessments were compared to calculate the accuracy, sensitivity, specificity and diagnostic accuracy of the ML models. The overall Top-1 accuracy of the ML model (39%) was lower than that of GPs (64%) and dermatologists (72%). When the analysis was limited to the diagnoses on which the algorithm had been explicitly trained (n = 82), the balanced Top-1 accuracy of the ML model increased (48%) and in the Top-3 (75%) was comparable to the GPs Top-3 accuracy (76%). The Top-5 accuracy of the ML model (89%) was comparable to the dermatologist Top-3 accuracy (90%). For the different diseases, the sensitivity of the model (Top-3 87% and Top-5 96%) is higher than that of the clinicians (Top-3 GPs 76% and Top-3 dermatologists 84%) only in the benign tumour pathology group, being on the other hand the most prevalent category (n = 53). About the satisfaction of professionals, 92% of the GPs considered it as a useful diagnostic support tool (DST) for the differential diagnosis and in 60% of the cases as an aid in the final diagnosis of the skin lesion. The overall diagnostic accuracy of the model in this study, under real-life conditions, is lower than that of both GPs and dermatologists. This result aligns with the findings of few existing prospective studies conducted under real-life conditions. The outcomes emphasize the significance of involving clinicians in the training of the model and the capability of ML models to assist GPs, particularly in differential diagnosis. Nevertheless, external testing in real-life conditions is crucial for data validation and regulation of these AI diagnostic models before they can be used in primary care.

Analyzing the Predictability of an Artificial Intelligence App (Tibot) in the Diagnosis of Dermatological Conditions: A Cross-sectional Study

BACKGROUND

Artificial intelligence (AI) aims to create programs that reproduce human cognition and processes involved in interpreting complex data. Dermatology relies on morphological features and is ideal for applying AI image recognition for assisted diagnosis. Tibot is an AI app that analyzes skin conditions and works on the principle of a convolutional neural network. Appropriate research analyzing the accuracy of such apps is necessary.

OBJECTIVE

This study aims to analyze the predictability of the Tibot AI app in the identification of dermatological diseases as compared to a dermatologist.

METHODS

This is a cross-sectional study. After taking informed consent, photographs of lesions of patients with different skin conditions were uploaded to the app. In every condition, the AI predicted three diagnoses based on probability, and these were compared with that by a dermatologist. The ability of the AI app to predict the actual diagnosis in the top one and top three anticipated diagnoses (prediction accuracy) was used to evaluate the app's effectiveness. Sensitivity, specificity, and positive predictive value were also used to assess the app's performance. Chi-square test was used to contrast categorical variables. P<.05 was considered statistically significant.

RESULTS

A total of 600 patients were included. Clinical conditions included alopecia, acne, eczema, immunological disorders, pigmentary disorders, psoriasis, infestation, tumors, and infections. In the anticipated top three diagnoses, the app's mean prediction accuracy was 96.1% (95% CI 94.3%-97.5%), while for the exact diagnosis, it was 80.6% (95% CI 77.2%-83.7%). The prediction accuracy (top one) for alopecia, acne, pigmentary disorders, and fungal infections was 97.7%, 91.7%, 88.5%, and 82.9%, respectively. Prediction accuracy (top three) for alopecia, eczema, and tumors was 100%. The sensitivity and specificity of the app were 97% (95% CI 95%-98%) and 98% (95% CI 98%-99%), respectively. There is a statistically significant association between clinical and AI-predicted diagnoses in all conditions (P<.001).

CONCLUSIONS

The AI app has shown promising results in diagnosing various dermatological conditions, and there is great potential for practical applicability.

Contemporary Advances in Computer-Assisted Bone Histomorphometry and Identification of Bone Cells in Culture

Static and dynamic bone histomorphometry and identification of bone cells in culture are labor-intensive and highly repetitive tasks. Several computer-assisted methods have been proposed to ease these tasks and to take advantage of the increased computational power available today. The present review aimed to provide an overview of contemporary methods utilizing specialized computer software to perform bone histomorphometry or identification of bone cells in culture. In addition, a brief historical perspective on bone histomorphometry is included. We identified ten publications using five different computer-assisted approaches (1) ImageJ and BoneJ; (2) Histomorph: OsteoidHisto, CalceinHisto, and TrapHisto; (3) Fiji/ImageJ2 and Trainable Weka Segmentation (TWS); (4) Visiopharm and artificial intelligence (AI); and (5) Osteoclast identification using deep learning with Single Shot Detection (SSD) architecture, Darknet and You Only Look Once (YOLO), or watershed algorithm (OC_Finder). The review also highlighted a substantial need for more validation studies that evaluate the accuracy of the new computational methods to the manual and conventional analyses of histological bone specimens and cells in culture using microscopy. However, a substantial evolution has occurred during the last decade to identify and separate bone cells and structures of interest. Most early studies have used simple image segmentation to separate structures of interest, whereas the most recent studies have utilized AI and deep learning. AI has been proposed to substantially decrease the amount of time needed for analyses and enable unbiased assessments. Despite the clear advantages of highly sophisticated computational methods, the limited nature of existing validation studies, particularly those that assess the accuracy of the third-generation methods compared to the second-generation methods, appears to be an important reason that these techniques have failed to gain wide acceptance.

Artificial intelligence in Dermatopathology

INTRODUCTION

Ever evolving research in medical field has reached an exciting stage with advent of newer technologies. With the introduction of digital microscopy, pathology has transitioned to become more digitally oriented speciality. The potential of artificial intelligence (AI) in dermatopathology is to aid the diagnosis, and it requires dermatopathologists' guidance for efficient functioning of artificial intelligence.

METHOD

Comprehensive literature search was performed using electronic online databases "PubMed" and "Google Scholar." Articles published in English language were considered for the review.

RESULTS

Convolutional neural network, a type of deep neural network, is considered as an ideal tool in image recognition, processing, classification, and segmentation. Implementation of AI in tumor pathology is involved in the diagnosis, grading, staging, and prognostic prediction as well as in identification of genetic or pathological features. In this review, we attempt to discuss the use of AI in dermatopathology, the attitude of patients and clinicians, its challenges, limitation, and potential opportunities in future implementation.

Artificial intelligence: Its role in dermatopathology

Artificial intelligence (AI), a major frontier in the field of medical research, can potentially lead to a paradigm shift in clinical practice. A type of artificial intelligence system known as convolutional neural network points to the possible utility of deep learning in dermatopathology. Though pathology has been traditionally restricted to microscopes and glass slides, recent advancement in digital pathological imaging has led to a transition making it a potential branch for the implementation of artificial intelligence. The current application of artificial intelligence in dermatopathology is to complement the diagnosis and requires a well-trained dermatopathologist's guidance for better designing and development of deep learning algorithms. Here we review the recent advances of artificial intelligence in dermatopathology, its applications in disease diagnosis and in research, along with its limitations and future potential.

Artificial intelligence in the detection of skin cancer

Recent advances in artificial intelligence (AI) in dermatology have demonstrated the potential to improve the accuracy of skin cancer detection. These capabilities may augment current diagnostic processes and improve the approach to the management of skin cancer. To explain this technology, we discuss fundamental terminology, potential benefits, and limitations of AI, and commercial applications relevant to dermatologists. A clear understanding of the technology may help to reduce physician concerns about AI and promote its use in the clinical setting. Ultimately, the development and validation of AI technologies, their approval by regulatory agencies, and widespread adoption by dermatologists and other clinicians may enhance patient care. Technology-augmented detection of skin cancer has the potential to improve quality of life, reduce health care costs by reducing unnecessary procedures, and promote greater access to high-quality skin assessment. Dermatologists play a critical role in the responsible development and deployment of AI capabilities applied to skin cancer.

Artificial Intelligence in Dermatology Image Analysis: Current Developments and Future Trends

BACKGROUND

Thanks to the rapid development of computer-based systems and deep-learning-based algorithms, artificial intelligence (AI) has long been integrated into the healthcare field. AI is also particularly helpful in image recognition, surgical assistance and basic research. Due to the unique nature of dermatology, AI-aided dermatological diagnosis based on image recognition has become a modern focus and future trend. Key scientific concepts of review: The use of 3D imaging systems allows clinicians to screen and label skin pigmented lesions and distributed disorders, which can provide an objective assessment and image documentation of lesion sites. Dermatoscopes combined with intelligent software help the dermatologist to easily correlate each close-up image with the corresponding marked lesion in the 3D body map. In addition, AI in the field of prosthetics can assist in the rehabilitation of patients and help to restore limb function after amputation in patients with skin tumors.

THE AIM OF THE STUDY

For the benefit of patients, dermatologists have an obligation to explore the opportunities, risks and limitations of AI applications. This study focuses on the application of emerging AI in dermatology to aid clinical diagnosis and treatment, analyzes the current state of the field and summarizes its future trends and prospects so as to help dermatologists realize the impact of new technological innovations on traditional practices so that they can embrace and use AI-based medical approaches more quickly.

Evaluation of an Artificial Intelligence web-based software to detect and classify dental structures and treatments in panoramic radiographs

OBJECTIVES

To evaluate the diagnostic reliability of a web-based Artificial Intelligence program on the detection and classification of dental structures and treatments present on panoramic radiographs.

METHODS

A total of 300 orthopantomographies (OPG) were randomly selected for this study. First, the images were visually evaluated by two calibrated operators with radiodiagnosis experience that, after consensus, established the "ground truth". Operators' findings on the radiographs were collected and classified as follows: metal restorations (MR), resin-based restorations (RR), endodontic treatment (ET), Crowns (C) and Implants (I). The orthopantomographies were then anonymously uploaded and automatically analyzed by the web-based software (Denti.Ai). Results were then stored, and a statistical analysis was performed by comparing them with the ground truth in terms of Sensitivity (S), Specificity (E), Positive Predictive Value (PPV) Negative Predictive Value (NPV) and its later representation in the area under (AUC) the Receiver Operating Characteristic (ROC) Curve.

RESULTS

Diagnostic metrics obtained for each study variable were as follows: (MR) S=85.48%, E=87.50%, PPV=82.8%, NPV=42.51%, AUC=0.869; (PR) S=41.11%, E=93.30%, PPV=90.24%, NPV=87.50%, AUC=0.672; (ET) S=91.9%, E=100%, PPV=100%, NPV=94.62%, AUC=0.960; (C) S=89.53%, E=95.79%, PPV=89.53%, NPV=95.79%, AUC=0.927; (I) S, E, PPV, NPV=100%, AUC=1.000.

CONCLUSIONS

Findings suggest that the web-based Artificial intelligence software provides a good performance on the detection of implants, crowns, metal fillings and endodontic treatments, not being so accurate on the classification of dental structures or resin-based restorations.

CLINICAL SIGNIFICANCE

General diagnostic and treatment decisions using orthopantomographies can be improved by using web-based artificial intelligence tools, avoiding subjectivity and lapses from the clinician.

Using Artificial Intelligence as a Diagnostic Decision Support Tool in Skin Disease: Protocol for an Observational Prospective Cohort Study

BACKGROUND

Dermatological conditions are a relevant health problem. Each person has an average of 1.6 skin diseases per year, and consultations for skin pathology represent 20% of the total annual visits to primary care and around 35% are referred to a dermatology specialist. Machine learning (ML) models can be a good tool to help primary care professionals, as it can analyze and optimize complex sets of data. In addition, ML models are increasingly being applied to dermatology as a diagnostic decision support tool using image analysis, especially for skin cancer detection and classification.

OBJECTIVE

This study aims to perform a prospective validation of an image analysis ML model as a diagnostic decision support tool for the diagnosis of dermatological conditions.

METHODS

In this prospective study, 100 consecutive patients who visit a participant general practitioner (GP) with a skin problem in central Catalonia were recruited. Data collection was planned to last 7 months. Anonymized pictures of skin diseases were taken and introduced to the ML model interface (capable of screening for 44 different skin diseases), which returned the top 5 diagnoses by probability. The same image was also sent as a teledermatology consultation following the current stablished workflow. The GP, ML model, and dermatologist's assessments will be compared to calculate the precision, sensitivity, specificity, and accuracy of the ML model. The results will be represented globally and individually for each skin disease class using a confusion matrix and one-versus-all methodology. The time taken to make the diagnosis will also be taken into consideration.

RESULTS

Patient recruitment began in June 2021 and lasted for 5 months. Currently, all patients have been recruited and the images have been shown to the GPs and dermatologists. The analysis of the results has already started.

CONCLUSIONS

This study will provide information about ML models' effectiveness and limitations. External testing is essential for regulating these diagnostic systems to deploy ML models in a primary care practice setting.

Using Artificial Intelligence as a Diagnostic Decision Support Tool in Skin Disease: Protocol for an Observational Prospective Cohort Study

BACKGROUND

Dermatological conditions are a relevant health problem. Each person has an average of 1.6 skin diseases per year, and consultations for skin pathology represent 20% of the total annual visits to primary care and around 35% are referred to a dermatology specialist. Machine learning (ML) models can be a good tool to help primary care professionals, as it can analyze and optimize complex sets of data. In addition, ML models are increasingly being applied to dermatology as a diagnostic decision support tool using image analysis, especially for skin cancer detection and classification.

OBJECTIVE

This study aims to perform a prospective validation of an image analysis ML model as a diagnostic decision support tool for the diagnosis of dermatological conditions.

METHODS

In this prospective study, 100 consecutive patients who visit a participant general practitioner (GP) with a skin problem in central Catalonia were recruited. Data collection was planned to last 7 months. Anonymized pictures of skin diseases were taken and introduced to the ML model interface (capable of screening for 44 different skin diseases), which returned the top 5 diagnoses by probability. The same image was also sent as a teledermatology consultation following the current stablished workflow. The GP, ML model, and dermatologist's assessments will be compared to calculate the precision, sensitivity, specificity, and accuracy of the ML model. The results will be represented globally and individually for each skin disease class using a confusion matrix and one-versus-all methodology. The time taken to make the diagnosis will also be taken into consideration.

RESULTS

Patient recruitment began in June 2021 and lasted for 5 months. Currently, all patients have been recruited and the images have been shown to the GPs and dermatologists. The analysis of the results has already started.

CONCLUSIONS

This study will provide information about ML models' effectiveness and limitations. External testing is essential for regulating these diagnostic systems to deploy ML models in a primary care practice setting.

Deep Learning in Dermatology: A Systematic Review of Current Approaches, Outcomes, and Limitations

Artificial intelligence (AI) has recently made great advances in image classification and malignancy prediction in the field of dermatology. However, understanding the applicability of AI in clinical dermatology practice remains challenging owing to the variability of models, image data, database characteristics, and variable outcome metrics. This systematic review aims to provide a comprehensive overview of dermatology literature using convolutional neural networks. Furthermore, the review summarizes the current landscape of image datasets, transfer learning approaches, challenges, and limitations within current AI literature and current regulatory pathways for approval of models as clinical decision support tools.

Telemedicine and e-Health in the Management of Psoriasis: Improving Patient Outcomes - A Narrative Review

The role of technology in dermatology is expanding. Telemedicine and eHealth are increasingly being used by doctors and patients in the management of psoriasis. This is a narrative review of the literature relating to the use of digital technology in the management of psoriasis. We divided psoriasis e-health into three areas: mobile phone applications, teledermatology and artificial Intelligence (AI). Literature searches were conducted using the following databases: Pubmed, Google Scholar, Scopus, both app stores using App Annie platform. The following words were used in searches; psoriasis, dermatology, mobile phone application, application, app, smartphone, teledermatology, telemedicine, artificial intelligence, AI, machine learning in various combinations. We defined three key questions, one relating to each of the 3 areas. We then reviewed the relevant papers found in the searches and selected the papers of highest research quality and greatest relevance in order to answer the questions. In addition, for apps, operating systems for IOS and android devices were searched for apps containing the key word "psoriasis" in the title using the app analytic website www.appannie.com on 08/11/21. Research publications linked to these apps were reviewed.

The Future of Precision Prevention for Advanced Melanoma

Precision prevention of advanced melanoma is fast becoming a realistic prospect, with personalized, holistic risk stratification allowing patients to be directed to an appropriate level of surveillance, ranging from skin self-examinations to regular total body photography with sequential digital dermoscopic imaging. This approach aims to address both underdiagnosis (a missed or delayed melanoma diagnosis) and overdiagnosis (the diagnosis and treatment of indolent lesions that would not have caused a problem). Holistic risk stratification considers several types of melanoma risk factors: clinical phenotype, comprehensive imaging-based phenotype, familial and polygenic risks. Artificial intelligence computer-aided diagnostics combines these risk factors to produce a personalized risk score, and can also assist in assessing the digital and molecular markers of individual lesions. However, to ensure uptake and efficient use of AI systems, researchers will need to carefully consider how best to incorporate privacy and standardization requirements, and above all address consumer trust concerns.

A machine learning algorithm for early detection of heel deep tissue injuries based on a daily history of sub-epidermal moisture measurements

Sub‐epidermal moisture is an established biophysical marker of pressure ulcer formation based on biocapacitance changes in affected soft tissues, which has been shown to facilitate early detection of these injuries. Artificial intelligence shows great promise in wound prevention and care, including in automated analyses of quantitative measures of tissue health such as sub‐epidermal moisture readings acquired over time for effective, patient‐specific, and anatomical‐site‐specific pressure ulcer prophylaxis. Here, we developed a novel machine learning algorithm for early detection of heel deep tissue injuries, which was trained using a database comprising six consecutive daily sub‐epidermal moisture measurements recorded from 173 patients in acute and post‐acute care settings. This algorithm was able to achieve strong predictive power in forecasting heel deep tissue injury events the next day, with sensitivity and specificity of 77% and 80%, respectively, revealing the clinical potential of artificial intelligence‐powered technology for hospital‐acquired pressure ulcer prevention. The current work forms the scientific basis for clinical implementation of machine learning algorithms that provide effective, early, and anatomy‐specific preventive interventions to minimise the occurrence of hospital‐acquired pressure ulcers based on routine tissue health status measurements.

Unbox the black-box for the medical explainable AI via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond

Explainable Artificial Intelligence (XAI) is an emerging research topic of machine learning aimed at unboxing how AI systems' black-box choices are made. This research field inspects the measures and models involved in decision-making and seeks solutions to explain them explicitly. Many of the machine learning algorithms cannot manifest how and why a decision has been cast. This is particularly true of the most popular deep neural network approaches currently in use. Consequently, our confidence in AI systems can be hindered by the lack of explainability in these black-box models. The XAI becomes more and more crucial for deep learning powered applications, especially for medical and healthcare studies, although in general these deep neural networks can return an arresting dividend in performance. The insufficient explainability and transparency in most existing AI systems can be one of the major reasons that successful implementation and integration of AI tools into routine clinical practice are uncommon. In this study, we first surveyed the current progress of XAI and in particular its advances in healthcare applications. We then introduced our solutions for XAI leveraging multi-modal and multi-centre data fusion, and subsequently validated in two showcases following real clinical scenarios. Comprehensive quantitative and qualitative analyses can prove the efficacy of our proposed XAI solutions, from which we can envisage successful applications in a broader range of clinical questions.

Determinants of Intention to Use Artificial Intelligence-Based Diagnosis Support System Among Prospective Physicians

Background: This study aimed to develop a theoretical model to explore the behavioral intentions of medical students to adopt an AI-based Diagnosis Support System.

Methods: This online cross-sectional survey used the unified theory of user acceptance of technology (UTAUT) to examine the intentions to use an AI-based Diagnosis Support System in 211 undergraduate medical students in Vietnam. Partial least squares (PLS) structural equational modeling was employed to assess the relationship between latent constructs.

Results: Effort expectancy (β = 0.201, p < 0.05) and social influence (β = 0.574, p < 0.05) were positively associated with initial trust, while no association was found between performance expectancy and initial trust (p > 0.05). Only social influence (β = 0.527, p < 0.05) was positively related to the behavioral intention.

Conclusions: This study highlights positive behavioral intentions in using an AI-based diagnosis support system among prospective Vietnamese physicians, as well as the effect of social influence on this choice. The development of AI-based competent curricula should be considered when reforming medical education in Vietnam.

The Digital Way to Intercept Psoriatic Arthritis

Psoriasis (PsO) and Psoriatic Arthritis (PsA) are chronic, immune-mediated diseases that share common etiopathogenetic pathways. Up to 30% of PsO patient may later develop PsA. In nearly 75% of cases, skin psoriatic lesions precede arthritic symptoms, typically 10 years prior to the onset of joint symptoms, while PsO diagnosis occurring after the onset of arthritis is described only in 15% of cases. Therefore, skin involvement offers to the rheumatologist a unique opportunity to study PsA in a very early phase, having a cohort of psoriatic “risk patients” that may develop the disease and may benefit from preventive treatment. Progression from PsO to PsA is often characterized by non-specific musculoskeletal symptoms, subclinical synovio-entheseal inflammation, and occasionally asymptomatic digital swelling such as painless toe dactylitis, that frequently go unnoticed, leading to diagnostic delay. The early diagnosis of PsA is crucial for initiating a treatment prior the development of significant and permanent joint damage. With the ongoing development of pharmacological treatments, early interception of PsA has become a priority, but many obstacles have been reported in daily routine. The introduction of digital technology in rheumatology may fill the gap in the physician-patient relationship, allowing more targeted monitoring of PsO patients. Digital technology includes telemedicine, virtual visits, electronic health record, wearable technology, mobile health, artificial intelligence, and machine learning. Overall, this digital revolution could lead to earlier PsA diagnosis, improved follow-up and disease control as well as maximizing the referral capacity of rheumatic centers.

A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI

Recently, artificial intelligence and machine learning in general have demonstrated remarkable performances in many tasks, from image processing to natural language processing, especially with the advent of deep learning (DL). Along with research progress, they have encroached upon many different fields and disciplines. Some of them require high level of accountability and thus transparency, for example, the medical sector. Explanations for machine decisions and predictions are thus needed to justify their reliability. This requires greater interpretability, which often means we need to understand the mechanism underlying the algorithms. Unfortunately, the blackbox nature of the DL is still unresolved, and many machine decisions are still poorly understood. We provide a review on interpretabilities suggested by different research works and categorize them. The different categories show different dimensions in interpretability research, from approaches that provide "obviously" interpretable information to the studies of complex patterns. By applying the same categorization to interpretability in medical research, it is hoped that: 1) clinicians and practitioners can subsequently approach these methods with caution; 2) insight into interpretability will be born with more considerations for medical practices; and 3) initiatives to push forward data-based, mathematically grounded, and technically grounded medical education are encouraged.

Discrimination Between Invasive and In Situ Melanomas Using Clinical Close-Up Images and a De Novo Convolutional Neural Network

Background: Melanomas are often easy to recognize clinically but determining whether a melanoma is in situ (MIS) or invasive is often more challenging even with the aid of dermoscopy. Recently, convolutional neural networks (CNNs) have made significant and rapid advances within dermatology image analysis. The aims of this investigation were to create a de novo CNN for differentiating between MIS and invasive melanomas based on clinical close-up images and to compare its performance on a test set to seven dermatologists.

Methods: A retrospective study including clinical images of MIS and invasive melanomas obtained from our department during a five-year time period (2016–2020) was conducted. Overall, 1,551 images [819 MIS (52.8%) and 732 invasive melanomas (47.2%)] were available. The images were randomized into three groups: training set (n = 1,051), validation set (n = 200), and test set (n = 300). A de novo CNN model with seven convolutional layers and a single dense layer was developed.

Results: The area under the curve was 0.72 for the CNN (95% CI 0.66–0.78) and 0.81 for dermatologists (95% CI 0.76–0.86) (P < 0.001). The CNN correctly classified 208 out of 300 lesions (69.3%) whereas the corresponding number for dermatologists was 216 (72.0%). When comparing the CNN performance to each individual reader, three dermatologists significantly outperformed the CNN.

Conclusions: For this classification problem, the CNN was outperformed by the dermatologist. However, since the algorithm was only trained and validated on 1,251 images, future refinement and development could make it useful for dermatologists in a real-world setting.

Insights into the Pathogenesis of HS and Therapeutical Approaches

Hidradenitis suppurativa (HS) is a debilitating, chronic, (auto)inflammatory disease primarily affecting apocrine gland-rich areas of the body. Although pathogenic mechanisms responsible for HS have not yet been fully elucidated, it is a multifactorial process whose main target is the terminal follicle. The role of the inflammatory process (and consequently of cytokine milieu) and of several other factors (genetics, lifestyle, hormonal status, microbiome, innate and adaptive immune systems) involved in HS pathogenesis has been investigated (and often defined) over the years with a view to transferring research results from bench to bedside and describing a unique and universally accepted pathogenetic model. This review will update readers on recent advances in our understanding of HS pathogenesis and novel (potential) medical therapies for patients with moderate-to-severe HS.

Artificial Intelligence in Dermatopathology: New Insights and Perspectives

In recent years, an increasing enthusiasm has been observed towards artificial intelligence and machine learning, involving different areas of medicine. Among these, although still in the embryonic stage, the dermatopathological field has also been partially involved, with the attempt to develop and train algorithms that could assist the pathologist in the differential diagnosis of complex melanocytic lesions. In this article, we face this new challenge of the modern era, carry out a review of the literature regarding the state of the art and try to determine promising future perspectives.

Artificial Intelligence-Aided Recognition of Pathological Characteristics and Subtype Classification of Superficial Perivascular Dermatitis

Background: Superficial perivascular dermatitis, an important type of inflammatory dermatosis, comprises various skin diseases, which are difficult to distinguish by clinical manifestations and need pathological imaging observation. Coupled with its complex pathological characteristics, the subtype classification depends to a great extent on dermatopathologists. There is an urgent need to develop an efficient approach to recognize the pathological characteristics and classify the subtypes of superficial perivascular dermatitis.

Methods: 3,954 pathological images (4 × and 10 ×) of three subtypes—psoriasiform, spongiotic and interface—of superficial perivascular dermatitis were captured from 327 cases diagnosed both clinically and pathologically. The control group comprised 1,337 pathological images of 85 normal skin tissue slides taken from the edge of benign epidermal cysts. First, senior dermatologists and dermatopathologists followed the structure–pattern analysis method to label the pathological characteristics that significantly contribute to classifying different subtypes on 4 × and 10 × images. A cascaded deep learning algorithm framework was then proposed to establish pixel-level pathological characteristics' masks and classify the subtypes by supervised learning.

Results: 13 different pathological characteristics were recognized, and the accuracy of subtype classification was 85.24%. In contrast, the accuracy of the subtype classification model without recognition was 71.35%.

Conclusion: Our cascaded deep learning model used small samples to deliver efficient recognition of pathological characteristics and subtype classification simultaneously. Moreover, the proposed method could be applied to both microscopic images and digital scanned images.

Current Challenges and Future Opportunities for XAI in Machine Learning-Based Clinical Decision Support Systems: A Systematic Review

Machine Learning and Artificial Intelligence (AI) more broadly have great immediate and future potential for transforming almost all aspects of medicine. However, in many applications, even outside medicine, a lack of transparency in AI applications has become increasingly problematic. This is particularly pronounced where users need to interpret the output of AI systems. Explainable AI (XAI) provides a rationale that allows users to understand why a system has produced a given output. The output can then be interpreted within a given context. One area that is in great need of XAI is that of Clinical Decision Support Systems (CDSSs). These systems support medical practitioners in their clinic decision-making and in the absence of explainability may lead to issues of under or over-reliance. Providing explanations for how recommendations are arrived at will allow practitioners to make more nuanced, and in some cases, life-saving decisions. The need for XAI in CDSS, and the medical field in general, is amplified by the need for ethical and fair decision-making and the fact that AI trained with historical data can be a reinforcement agent of historical actions and biases that should be uncovered. We performed a systematic literature review of work to-date in the application of XAI in CDSS. Tabular data processing XAI-enabled systems are the most common, while XAI-enabled CDSS for text analysis are the least common in literature. There is more interest in developers for the provision of local explanations, while there was almost a balance between post-hoc and ante-hoc explanations, as well as between model-specific and model-agnostic techniques. Studies reported benefits of the use of XAI such as the fact that it could enhance decision confidence for clinicians, or generate the hypothesis about causality, which ultimately leads to increased trustworthiness and acceptability of the system and potential for its incorporation in the clinical workflow. However, we found an overall distinct lack of application of XAI in the context of CDSS and, in particular, a lack of user studies exploring the needs of clinicians. We propose some guidelines for the implementation of XAI in CDSS and explore some opportunities, challenges, and future research needs.