In recent years, there has been a significant improvement in the accuracy of the classification of pigmented skin lesions using artificial intelligence algorithms. Intelligent analysis and classification systems are significantly superior to visual diagnostic methods used by dermatologists and oncologists. However, the application of such systems in clinical practice is severely limited due to a lack of generalizability and risks of potential misclassification. Successful implementation of artificial intelligence-based tools into clinicopathological practice requires a comprehensive study of the effectiveness and performance of existing models, as well as further promising areas for potential research development. The purpose of this systematic review is to investigate and evaluate the accuracy of artificial intelligence technologies for detecting malignant forms of pigmented skin lesions. For the study, 10,589 scientific research and review articles were selected from electronic scientific publishers, of which 171 articles were included in the presented systematic review. All selected scientific articles are distributed according to the proposed neural network algorithms from machine learning to multimodal intelligent architectures and are described in the corresponding sections of the manuscript. This research aims to explore automated skin cancer recognition systems, from simple machine learning algorithms to multimodal ensemble systems based on advanced encoder-decoder models, visual transformers (ViT), and generative and spiking neural networks. In addition, as a result of the analysis, future directions of research, prospects, and potential for further development of automated neural network systems for classifying pigmented skin lesions are discussed.

The early identification of rotten potatoes is one of the most important challenges in a storage facility because of the inconspicuous symptoms of rot, the high density of storage, and environmental factors (such as temperature, humidity, and ambient gases). An electronic nose system based on an ensemble convolutional neural network (ECNN, a powerful feature extraction method) was developed to detect potatoes with different degrees of rot. Three types of potatoes were detected: normal samples, slightly rotten samples, and totally rotten samples. A feature discretization method was proposed to optimize the impact of ambient gases on electronic nose signals by eliminating redundant information from the features. The ECNN based on original features presented good results for the prediction of rotten potatoes in both laboratory and storage environments, and the accuracy of the prediction results was 94.70% and 90.76%, respectively. Moreover, the application of the feature discretization method significantly improved the prediction results, and the accuracy of prediction results improved by 1.59% and 3.73%, respectively. Above all, the electronic nose system performed well in the identification of three types of potatoes by using the ECNN, and the proposed feature discretization method was helpful in reducing the interference of ambient gases.

OBJECTIVE: Left ventricular hypertrophy (LVH) can impair ejection function and elevate the risk of heart failure. Therefore, early detection through screening is crucial. This study aimed to propose a novel method to enhance LVH detection using 12-lead electrocardiogram (ECG) waveforms with a two-dimensional (2D) convolutional neural network (CNN).
METHODS: Utilizing 42,127 pairs of ECG-transthoracic echocardiogram data, we pre-processed raw data into single-shot images derived from each ECG lead and conducted lead selection to optimize LVH diagnosis. Our proposed one-shot screening method, implemented during pre-processing, enables the superimposition of waveform source data of any length onto a single-frame image, thereby addressing the limitations of the one-dimensional (1D) approach. We developed a deep learning model with a 2D-CNN structure and machine learning models for LVH detection. To assess our method, we also compared our results with conventional ECG criteria and those of a prior study that used a 1D-CNN approach, utilizing the same dataset from the University of Tokyo Hospital for LVH diagnosis.
RESULTS: For LVH detection, the average area under the receiver operating characteristic curve (AUROC) was 0.916 for the 2D-CNN model, which was significantly higher than that obtained using logistic regression and random forest methods, as well as the two conventional ECG criteria (AUROC of 0.766, 0.790, 0.599, and 0.622, respectively). Incorporating additional metadata, such as ECG measurement data, further improved the average AUROC to 0.921. The model\'s performance remained stable across two different annotation criteria and demonstrated significant superiority over the performance of the 1D-CNN model used in a previous study (AUROC of 0.807).
CONCLUSIONS: This study introduces a robust and computationally efficient method that outperforms 1D-CNN models utilized in previous studies for LVH detection. Our method can transform waveforms of any length into fixed-size images and leverage the selected lead of the ECG, ensuring adaptability in environments with limited computational resources. The proposed method holds promise for integration into clinical practice as a tool for early diagnosis, potentially enhancing patient outcomes by facilitating earlier treatment and management.

This paper reports on research to design an ensemble deep learning framework that integrates fine-tuned, three-stream hybrid deep neural network (i.e., Ensemble Deep Learning Model, EDLM), employing Convolutional Neural Network (CNN) to extract facial image features, detect and accurately classify the pain. To develop the approach, the VGGFace is fine-tuned and integrated with Principal Component Analysis and employed to extract features in images from the Multimodal Intensity Pain database at the early phase of the model fusion. Subsequently, a late fusion, three layers hybrid CNN and recurrent neural network algorithm is developed with their outputs merged to produce image-classified features to classify pain levels. The EDLM model is then benchmarked by means of a single-stream deep learning model including several competing models based on deep learning methods. The results obtained indicate that the proposed framework is able to outperform the competing methods, applied in a multi-level pain detection database to produce a feature classification accuracy that exceeds 89 %, with a receiver operating characteristic of 93 %. To evaluate the generalization of the proposed EDLM model, the UNBC-McMaster Shoulder Pain dataset is used as a test dataset for all of the modelling experiments, which reveals the efficacy of the proposed method for pain classification from facial images. The study concludes that the proposed EDLM model can accurately classify pain and generate multi-class pain levels for potential applications in the medical informatics area, and should therefore, be explored further in expert systems for detecting and classifying the pain intensity of patients, and automatically evaluating the patients\' pain level accurately.

Recent developments in artificial intelligence have generated increasing interest to deploy automated image analysis for diagnostic imaging and large-scale clinical applications. However, inaccuracy from automated methods could lead to incorrect conclusions, diagnoses or even harm to patients. Manual inspection for potential inaccuracies is labor-intensive and time-consuming, hampering progress towards fast and accurate clinical reporting in high volumes. To promote reliable fully-automated image analysis, we propose a quality control-driven (QCD) segmentation framework. It is an ensemble of neural networks that integrate image analysis and quality control. The novelty of this framework is the selection of the most optimal segmentation based on predicted segmentation accuracy, on-the-fly. Additionally, this framework visualizes segmentation agreement to provide traceability of the quality control process. In this work, we demonstrated the utility of the framework in cardiovascular magnetic resonance T1-mapping - a quantitative technique for myocardial tissue characterization. The framework achieved near-perfect agreement with expert image analysts in estimating myocardial T1 value (r=0.987,p<.0005; mean absolute error (MAE)=11.3ms), with accurate segmentation quality prediction (Dice coefficient prediction MAE=0.0339) and classification (accuracy=0.99), and a fast average processing time of 0.39 second/image. In summary, the QCD framework can generate high-throughput automated image analysis with speed and accuracy that is highly desirable for large-scale clinical applications.

Lung cancer is the number one cause of cancer-related deaths in the United States as well as worldwide. Radiologists and physicians experience heavy daily workloads, thus are at high risk for burn-out. To alleviate this burden, this narrative literature review compares the performance of four different artificial intelligence (AI) models in lung nodule cancer detection, as well as their performance to physicians/radiologists reading accuracy. A total of 648 articles were selected by two experienced physicians with over 10 years of experience in the fields of pulmonary critical care, and hospital medicine. The data bases used to search and select the articles are PubMed/MEDLINE, EMBASE, Cochrane library, Google Scholar, Web of science, IEEEXplore, and DBLP. The articles selected range from the years between 2008 and 2019. Four out of 648 articles were selected using the following inclusion criteria: 1) 18-65 years old, 2) CT chest scans, 2) lung nodule, 3) lung cancer, 3) deep learning, 4) ensemble and 5) classic methods. The exclusion criteria used in this narrative review include: 1) age greater than 65 years old, 2) positron emission tomography (PET) hybrid scans, 3) chest X-ray (CXR) and 4) genomics. The model performance outcomes metrics are measured and evaluated in sensitivity, specificity, accuracy, receiver operator characteristic (ROC) curve, and the area under the curve (AUC). This hybrid deep-learning model is a state-of-the-art architecture, with high-performance accuracy and low false-positive results. Future studies, comparing each model accuracy at depth is key. Automated physician-assist systems as this model in this review article help preserve a quality doctor-patient relationship.

In our earlier work, we have demonstrated that it is possible to characterize binary mixtures using single component descriptors by applying various mixing rules. We also showed that these methods were successful in building predictive QSPR models to study various mixture properties of interest. Here in, we developed a QSPR model of an excess thermodynamic property of binary mixtures i.e. excess molar volume (V(E) ). In the present study, we use a set of mixture descriptors which we earlier designed to specifically account for intermolecular interactions between the components of a mixture and applied successfully to the prediction of infinite-dilution activity coefficients using neural networks (part 1 of this series). We obtain a significant QSPR model for the prediction of excess molar volume (V(E) ) using consensus neural networks and five mixture descriptors. We find that hydrogen bond and thermodynamic descriptors are the most important in determining excess molar volume (V(E) ), which is in line with the theory of intermolecular forces governing excess mixture properties. The results also suggest that the mixture descriptors utilized herein may be sufficient to model a wide variety of properties of binary and possibly even more complex mixtures.