Recently, skin cancer is one of the spread and dangerous cancers around the world. Early detection of skin cancer can reduce mortality. Traditional methods for skin cancer detection are painful, time-consuming, expensive, and may cause the disease to spread out. Dermoscopy is used for noninvasive diagnosis of skin cancer. Artificial Intelligence (AI) plays a vital role in diseases’ diagnosis especially in biomedical engineering field. The automated detection systems based on AI reduce the complications in the traditional methods and can improve skin cancer’s diagnosis rate. In this paper, automated early detection system for skin cancer dermoscopic images using artificial intelligent is presented. Adaptive snake (AS) and region growing (RG) algorithms are used for automated segmentation and compared with each other. The results show that AS is accurate and efficient (accuracy = 96%) more than RG algorithm (accuracy = 90%). Artificial Neural networks (ANN) and support vector machine (SVM) algorithms are used for automated classification compared with each other. The proposed system with ANN algorithm shows high accuracy (94%), precision (96%), specificity (95.83%), sensitivity (recall) (92.30%), and F1-score (0.94). The proposed system is easy to use, time consuming, enables patients to make early detection for skin cancer and has high efficiency.

Fetal health is a vital aspect of pregnancy, influencing both the mother and her fetus. Frequent observation and prompt response are essential for achieving optimal outcomes. It is important to assess fetal health within the womb, ensuring that any potential issues are addressed rapidly. Prioritizing fetal monitoring is essential for safe and healthy pregnancy, one of such methods is Cardiotocography (CTG). CTG is employed to monitor the uterine contraction patterns and fetal heart rate during pregnancy and labor. The aim of this paper is to use artificial intelligence to enhance the accuracy of fetal health prediction and enhance clinical decision-making. Seven machine learning (ML) algorithms and five deep learning (DL) algorithms are applied. In addition, H₂O.ai and Lazy predict platforms were applied for prediction. Ensemble learning was employed to combine the most effective models to construct the Blender model, emulate the traditional ML, DL models, and ML with DL in meta classifiers. The results for ML models showed that meta-model with stacking classifier had the highest accuracy of 98.9%. The results for DL models showed that ANN had the highest accuracy of 97.7%. The analysis of each model’s performance demonstrated that the proposed stacking classifier achieved 98.9% accuracy, 99% precision, 98.6% recall, 99.3% F1-score, and 99.8% area under the ROC curve. This implies that stacking classifier model demonstrates a strong capability in predicting fetal health and it can be integrated with the CTG device for real-time monitoring and medical follow-up by healthcare providers.

Recently, vertebral tumor prediction using image-based vertebral bio-mathematical modeling with accurate segmentation has been considered as a significant area of research. Further, precise lumbar spinal segmentation is essential to the clinicians for tumor analysis. Therefore, precise and reliable segmentation process is required to help radiologists and doctors to identify different vertebrae tumors with better prediction ratio. The exact vertebral disks segmentation of spinal bones from medical images is a complex process in dealing with different deformities and pathologies in accordance to the conventional techniques such as Deep Convolutional Neural Network (DCNN), Finite Element Analysis (FEA), Principal Component and Factor Analysis (PCFA), Multi-Parameter Ensemble Learning (MPEL), Hierarchical Conditional Random Forest (HCRF), and Deep Siamese Neural Network (DSNN). Therefore, to overcome the present drawbacks, Analytical Transform Assisted Statistical Characteristic Decomposition Model (ATS-CDM) is proposed in this paper for the accurate prediction of vertebral tumor detection and segmentation. This technique is used for the calibration of the segmentation procedures in vertebral tumor image prediction and Receiver Operating Curve (ROC) grading for the lumbar spines. The significant objective of the model-fitting algorithm iterates the tumor regions and measures the current region's variance for the accurate identification of tumor. The outcomes show potential and promising results at lab scale evaluation through analyzing the vertebral datasets with Intra-Discal Pressure (IDP) images for experimental validation with 98.7% bending and 98.88 % segmentation accuracy leads to 94.2 ± 0.2 % to 97.02 ± 0.2 % average ROC grading.

Parkinson’s disease (PD) is one of the most prevalent neurodegenerative disorders, with a sharp increase predicted. Classifying and predicting PD at an early stage is crucial. Application of artificial intelligence (AI) is a significant factor in the diagnosis of various disorders. Based on patient data, machine learning (ML) and deep learning (DL) can automatically predict PD. This research aims to develop an automated approach for early PD prediction based on vocal symptoms and AI techniques. To forecast PD, specific AI models have been implemented. Extreme gradient boosting (XGB or XGBoost), artificial neural networks (ANN), Naive Bayes (NB), K-nearest neighbor (KNN), multilayer perceptron ( MLP), logistic regression (LR), support vector machine (SVM), and ridge classifier with cross-validation (RidgeCV) were among the AI models used. The dataset was subjected to various data preprocessing approaches, such as Min–Max scaling and synthetic minority over-sampling technique (SMOTE). Sensitivity, accuracy, F1-score, precision, specificity, and the area under the receiver operating characteristic (ROC) curve (AUC) were among the evaluation measures used to assess the effectiveness of the implemented AI system. The results demonstrated that, with 98% accuracy, 97% precision, 100% sensitivity, 98% F1-score, 97% specificity, and 100% AUC, the XGB model utilizing SMOTE approach achieved the best results. With the proposed approach, patients can forecast their PD early. The proposed work contributes significantly to the field of neurodegenerative disease research by demonstrating the effectiveness of AI techniques in early PD prediction, which can have profound implications for patient care and treatment strategies.

The automatic diagnosis of hyperkinetic dysphonia using deep learning techniques represents a significant advancement in voice disorder assessment. This approach leverages the analysis of spectrograms, Mel-spectrograms, and Mel-Frequency Cepstral Coefficients (MFCCs) extracted from speech recordings to enable precise diagnosis. Hyperkinetic dysphonia is a voice disorder characterized by excessive vocal fold tension, traditionally diagnosed through subjective assessments and invasive procedures. These conventional methods are often time-consuming, uncomfortable for patients, and prone to inconsistencies, which can lead to potential delays in treatment. In this study, various deep learning models, including multilayer perceptron (MLP), convolutional neural networks (CNN), and long short-term memory (LSTM), a specialized type of recurrent neural network (RNN) are employed to offer a non-invasive, efficient, and highly accurate diagnostic alternative. The analysis is conducted on healthy and hyperkinetic voice samples from the VOICED dataset. Voice signals are processed and segmented into five and ten segments, followed by the extraction of spectrograms, Mel-spectrograms, and MFCCs for each segment. Each model was trained and evaluated using the bootstrapping technique for robustness. The classification results demonstrate that the accuracy of all models ranges from approximately 67.1% to 99.8%. The regularized MLP model, which utilized Mel-spectrogram features extracted from 10 audio segments, exhibits superior performance across six metrics: accuracy (0.9984), precision (0.9985), F1 score (0.9984), recall (0.9984), specificity (0.9980), and Area Under the Curve (AUC) (0.9999). It outperforms existing methods and other deep learning approaches. These findings emphasize the potential of integrating deep learning models into clinical practice for diagnosing hyperkinetic dysphonia. The proposed approach provides a non-invasive, reliable, and highly accurate diagnostic tool, reducing the workload of healthcare professionals while enhancing patient care outcomes.