Objective: Skin diseases constitute a widespread health concern, and the application of machine learning and deep learning algorithms has been instrumental in improving diagnostic accuracy and treatment effectiveness. This paper aims to provide a comprehensive review of the existing research on the utilization of machine learning and deep learning in the field of skin disease diagnosis, with a particular focus on recent widely used methods of deep learning. The present challenges and constraints were also analyzed and possible solutions were proposed. Methods: We collected comprehensive works from the literature, sourced from distinguished databases including IEEE, Springer, Web of Science, and PubMed, with a particular emphasis on the most recent 5-year advancements. From the extensive corpus of available research, twenty-nine articles relevant to the segmentation of dermatological images and forty-five articles about the classification of dermatological images were incorporated into this review. These articles were systematically categorized into two classes based on the computational algorithms utilized: traditional machine learning algorithms and deep learning algorithms. An in-depth comparative analysis was carried out, based on the employed methodologies and their corresponding outcomes. Conclusions: Present outcomes of research highlight the enhanced effectiveness of deep learning methods over traditional machine learning techniques in the field of dermatological diagnosis. Nevertheless, there remains significant scope for improvement, especially in improving the accuracy of algorithms. The challenges associated with the availability of diverse datasets, the generalizability of segmentation and classification models, and the interpretability of models also continue to be pressing issues. Moreover, the focus of future research should be appropriately shifted. A significant amount of existing research is primarily focused on melanoma, and consequently there is a need to broaden the field of pigmented dermatology research in the future. These insights not only emphasize the potential of deep learning in dermatological diagnosis but also highlight directions that should be focused on.

The choice of an electroencephalograph (EEG) reference is a practical issue for the study of brain functional connectivity. To study how EEG reference influence functional connectivity estimation (FCE), this study compares the differences of FCE resulting from the different references such as REST (the reference electrode standardization technique), average reference (AR), linked mastoids (LM), and left mastoid references (LR). Simulations involve two parts. One is based on 300 dipolar pairs, which are located on the superficial cortex with a radial source direction. The other part is based on 20 dipolar pairs. In each pair, the dipoles have various orientation combinations. The relative error (RE) and Hamming distance (HD) between functional connectivity matrices of ideal recordings and that of recordings obtained with different references, are metrics to compare the differences of the scalp functional connectivity graph (FCG) derived from those two kinds of recordings. Lower RE and HD values imply more similarity between the two FCGs. Using the ideal recording (IR) as a standard, the results show that AR, LM and LR perform well only in specific conditions, i.e., AR performs stable when there is no upward component in sources’ orientation. LR achieves desirable results when the sources’ locations are away from left ear. LM achieves an indistinct difference with IR, i.e., when the distribution of source locations is symmetric along the line linking the two ears. However, REST not only achieves excellent performance for superficial and radial dipolar sources, but also achieves a stable and robust performance with variable source locations and orientations. Benefitting from the stable and robust performance of REST versus other reference methods, REST might best recover the real FCG of EEG. Thus, REST based FCG may be a good candidate to compare the FCG of EEG based on different references from different labs.

Vitiligo is a hypopigmented skin disease characterized by the loss of melanin. The progressive nature and widespread incidence of vitiligo necessitate timely and accurate detection. Usually, a single diagnostic test often falls short of providing definitive confirmation of the condition, necessitating the assessment by dermatologists who specialize in vitiligo. However, the current scarcity of such specialized medical professionals presents a significant challenge. To mitigate this issue and enhance diagnostic accuracy, it is essential to build deep learning models that can support and expedite the detection process. This study endeavors to establish a deep learning framework to enhance the diagnostic accuracy of vitiligo. To this end, a comparative analysis of five models including ResNet (ResNet34, ResNet50, and ResNet101 models) and Swin Transformer series (Swin Transformer Base, and Swin Transformer Large models), were conducted under the uniform condition to identify the model with superior classification capabilities. Moreover, the study sought to augment the interpretability of these models by selecting one that not only provides accurate diagnostic outcomes but also offers visual cues highlighting the regions pertinent to vitiligo. The empirical findings reveal that the Swin Transformer Large model achieved the best performance in classification, whose AUC, accuracy, sensitivity, and specificity are 0.94, 93.82%, 94.02%, and 93.5%, respectively. In terms of interpretability, the highlighted regions in the class activation map correspond to the lesion regions of the vitiligo images, which shows that it effectively indicates the specific category regions associated with the decision-making of dermatological diagnosis. Additionally, the visualization of feature maps generated in the middle layer of the deep learning model provides insights into the internal mechanisms of the model, which is valuable for improving the interpretability of the model, tuning performance, and enhancing clinical applicability. The outcomes of this study underscore the significant potential of deep learning models to revolutionize medical diagnosis by improving diagnostic accuracy and operational efficiency. The research highlights the necessity for ongoing exploration in this domain to fully leverage the capabilities of deep learning technologies in medical diagnostics.

Depression (DP) and schizophrenia (SCZ) are both highly prevalent psychiatric disorders, and their diagnosis depends on the examination of symptoms and clinical tests, which can be subjective. As a measure of real-time neural activity, Electroencephalographic (EEG) has shown its usability to classify people either as normal or as having DP or SCZ, but automatic classification between the three categories (DP, SCZ and the normal) was rarely reported. Here, we propose an automatic diagnostic framework based on a convolutional neural network called the Multi-Channel Frequency Network (MUCHf-Net), which automatically learns feature representations of EEGs that characterize them as normal, DP, or SCZ. Two EEG databases were used in this study, the first one contains EEGs from 300 individuals (DP: 100, SCZ: 100, normal: 100) collecting from our hospital, and the second contains EEGs from 30 individuals (DP: 10, SCZ: 10, normal: 10) from public available datasets, and the spectrum matrices from these multi-channel EEGs were feed into MUCHf-Net. The results showed that: (1) MUCHf-Net accurately distinguished normal EEGs from DP or SCZ EEGs (accuracy: 91.12%; F1 score: 0.8947); (2) low-frequency bands (delta, theta, alpha) contributed the most important information to the classification model; (3) features located in the frontal and parietal lobes contributed more than other regions did; (4) MUCHf-Net fine-tuned on public datasets also had high classification accuracy: 87.71% (triple: normal, SCZ or DP) and 79.27% (binary: psychiatric disorders (DP or SCZ) or normal). Our study shows that deep learning has the potential to become an important tool for assisting in the diagnosis of psychiatric disorders.

Correct lateralization of temporal lobe epilepsy (TLE) is critical for improving surgical outcomes. As a relatively new noninvasive clinical recording system, magnetoencephalography (MEG) has rarely been applied for determining lateralization of unilateral TLE. Here we propose a framework for using resting‐state brain‐network features and support vector machine (SVM) for TLE lateralization based on MEG. We recruited 15 patients with left TLE, 15 patients with right TLE, and 15 age‐ and sex‐matched healthy controls. The lateralization problem was then transferred into a series of binary classification problems, including left TLE versus healthy control, right TLE versus healthy control, and left TLE versus right TLE. Brain‐network features were extracted for each participant using three network metrics (nodal degree, betweenness centrality, and nodal efficiency). A radial basis function kernel SVM (RBF‐SVM) was employed as the classifier. The leave‐one‐subject‐out cross‐validation strategy was used to test the ability of this approach to overcome individual differences. The results revealed that the nodal degree performed best for left TLE versus healthy control and right TLE versus healthy control, with accuracy of 80.76% and 75.00%, respectively. Betweenness centrality performed best for left TLE versus right TLE with an accuracy of 88.10%. The proposed approach demonstrated that MEG is a good candidate for solving the lateralization problem in unilateral TLE using various brain‐network features.