Single-molecule detection (SMD) has reformed analytical science by enabling the direct observation of individual molecular events, thus overcoming the limitations of ensemble-averaged measurements. This review presents a comprehensive analysis of the principles, devices, and emerging materials that have shaped the current landscape of SMD. We explore a wide range of sensing mechanisms, including surface plasmon resonance, mechanochemical transduction, transistor-based sensing, optical microfiber platforms, fluorescence-based techniques, Raman scattering, and recognition tunneling, which offer distinct advantages in terms of label-free operation, ultrasensitivity, and real-time responsiveness. Each technique is critically examined through representative case studies, revealing how innovations in device architecture and signal amplification strategies have collectively pushed the detection limits into the femtomolar to attomolar range. Beyond the sensing principles, this review highlights the transformative role of advanced nanomaterials such as graphene, carbon nanotubes, quantum dots, MnO₂ nanosheets, upconversion nanocrystals, and magnetic nanoparticles. These materials enable new transduction pathways and augment the signal strength, specificity, and integration into compact and wearable biosensing platforms. We also detail the multifaceted applications of SMD across biomedical diagnostics, environmental monitoring, food safety, neuroscience, materials science, and quantum technologies, underscoring its relevance to global health, safety, and sustainability. Despite significant progress, the field faces several critical challenges, including signal reproducibility, biocompatibility, fabrication scalability, and data interpretation complexity. To address these barriers, we propose future research directions involving multimodal transduction, AI-assisted signal analytics, surface passivation techniques, and modular system design for field-deployable diagnostics. By providing a cross-disciplinary synthesis of device physics, materials science, and real-world applications, this review offers a comprehensive roadmap for the next generation of SMD technologies, poised to impact both fundamental research and translational healthcare.

Artificial intelligence (AI) and machine learning (ML) are transforming medical diagnostics, but human nail, an easily accessible and rich biological substrate, is still not fully exploited in the digital health field. Nail pathologies are easily diagnosed, non-invasive disease biomarkers, including systemic diseases such as anemia, diabetes, psoriasis, melanoma, and fungal diseases. This review presents the first big synthesis of image analysis for nail lesions incorporating AI/ML for diagnostic purposes. Where dermatological reviews to date have been more wide-ranging in scope, our review will focus specifically on diagnosis and screening related to nails. The various technological modalities involved (smartphone imaging, dermoscopy, Optical Coherence Tomography) will be presented, together with the different processing techniques for images (color corrections, segmentation, cropping of regions of interest), and models that range from classical methods to deep learning, with annotated descriptions of each. There will also be additional descriptions of AI applications related to some diseases, together with analytical discussions regarding real-world impediments to clinical application, including scarcity of data, variations in skin type, annotation errors, and other laws of clinical adoption. Some emerging solutions will also be emphasized: explainable AI (XAI), federated learning, and platform diagnostics allied with smartphones. Bridging the gap between clinical dermatology, artificial intelligence and mobile health, this review consolidates our existing knowledge and charts a path through yet others to scalable, equitable, and trustworthy nail based medically diagnostic techniques. Our findings advocate for interdisciplinary innovation to bring AI-enabled nail analysis from lab prototypes to routine healthcare and global screening initiatives.

Detection of skin cancer has become an increasingly prevalent health issue for which there is a need for reliable methods of detection to improve patient outcomes, minimize delays in diagnosis, and ensure clinical referral. Advances in computer vision and artificial intelligence have allowed automatic analyses of dermoscopic, clinical, and smartphone imaging of skin lesions. While numerous models have been found to perform very well on the basis of curated benchmark datasets, their accuracy in practical settings still needs to be evaluated. This review critically evaluates the literature from 2015 to 2025. Rather than considering the Dice, Jaccard, AUC, sensitivity, and specificity metrics on an absolute basis, this review evaluates them relative to dataset quality, validation process, external testing, statistical analysis, and risk of bias. Key areas of focus include dataset imbalance, lack of coverage of darker skin, poor external validation, explainability, uncertainty quantification, and barriers to clinical adoption. This review provides valuable insights for researchers, practitioners, dataset producers, and healthcare providers involved in developing AI-enabled dermatology tools.

Deep learning-based automated classification of dermatoscopic skin lesions has exhibited promising potential in diagnostics. However, two prominent issues need to be addressed before achieving high-quality diagnostic tools: inconsistent performance in the case of imbalanced classes and poor clinical interpretability of models. Even though some studies have attempted to leverage both deep and shallow learning by combining pretrained convolutional neural networks (CNNs)-based feature extraction with classical machine learning (ML) models, very few of them systematically explore several model combinations based on various clinically important metrics, such as F1-score, precision, recall, accuracy, etc., and utilize decision threshold calibration techniques. In this research, we present an evaluation of a systematic framework with threshold calibration for the comparison of several hybrid models on seven-class skin lesion classification (multi-class) on the HAM10000 dataset. In particular, we used deep features extracted from three pretrained CNN architectures, i.e., DenseNet201, InceptionV3 and EfficientNet-B4. These deep features were used as inputs for six different classical classifiers. As a result, we obtained 18 comparable hybrid models that were then systematically compared by multiple clinically relevant metrics: accuracy, macro-precision, macro-recall, macro-F1, ROC-AUC, Precision-Recall-AUC, and log loss. Also, fold-wise optimization of decision thresholds was performed, which was based on the maximization of the macro-F1 score. Finally, we found out that DenseNet201 with an SVM-RBF classifier yielded the highest performance among all 18 tested models, showing 90.88% accuracy, 90.7% macro-precision, and 0.921 ROC-AUC. To analyze the clinical plausibility, top-performing models were further explained with explainable artificial intelligence (XAI) techniques: Grad-CAM, LIME and Occlusion Sensitivity. Results show that the most successful models concentrated mostly on lesion-specific areas. Overall, this study contributes a reproducible hybrid-XAI model-selection framework rather than a single black-box classifier, supporting more transparent and clinically meaningful skin lesion diagnosis.