Automatic vision-based inspection systems have played a key role in product quality assessment for decades through the segmentation, detection, and classification of defects. Historically, machine learning frameworks, based on hand-crafted feature extraction, selection, and validation, counted on a combined approach of parameterized image processing algorithms and explicated human knowledge. The outstanding performance of deep learning (DL) for vision systems, in automatically discovering a feature representation suitable for the corresponding task, has exponentially increased the number of scientific articles and commercial products aiming at industrial quality assessment. In such a context, this article reviews more than 220 relevant articles from the related literature published until February 2023, covering the recent consolidation and advances in the field of fully-automatic DL-based surface defects inspection systems, deployed in various industrial applications. The analyzed papers have been classified according to a bi-dimensional taxonomy, that considers both the specific defect recognition task and the employed learning paradigm. The dependency on large and high-quality labeled datasets and the different neural architectures employed to achieve an overall perception of both well-visible and subtle defects, through the supervision of fine or/and coarse data annotations have been assessed. The results of our analysis highlight a growing research interest in defect representation power enrichment, especially by transferring pre-trained layers to an optimized network and by explaining the network decisions to suggest trustworthy retention or rejection of the products being evaluated.

Patients with Parkinson's disease (PD) may have difficulties in speaking because of reduced coordination of the muscles that control breathing, phonation, articulation, and prosody. Symptoms that may occur are weakening of the volume of the voice, voice monotony, changes in the quality of the voice, the speed of speech, uncontrolled repetition of words, and difficult speech intelligibility. To date, evaluation of the speech intelligibility is performed based on the unified PD rating scale. Specifically, section 3.1 (eloquence) of the cited scale provides the specialist with some tips to evaluate the patient's speech ability. With the aim of evaluating the speech intelligibility by measuring the variation in parameters in an objective manner, we show that a speech-to-text (STT) system could help specialists to obtain an accurate and objective measure of speech, phrase, and word intelligibility in PD. STT systems are based on methodologies and technologies that enable the recognition and translation of spoken language into text by computers and computerized devices. We decided to base our study on Google STT conversion. We expand Voxtester, a software system for digital assessment of voice and speech changes in PD, in order to perform this study. No previous studies have been presented to address the mentioned challenges based on STT. The experiments here presented are related with detection/classification between pathological speech from patients with PD and regular speech from healthy control group. The results are very interesting and are an important step toward assessing the intelligibility of the speech of PD patients.

Previous studies have shown abnormal power and functional connectivity of resting state electroencephalographic (EEG) rhythms in groups of Alzheimer’s disease (AD) compared to healthy elderly (Nold) subjects. Here we tested the best classification rate of 120 AD patients and 100 matched Nold subjects using EEG markers based on cortical sources of power and functional connectivity of these rhythms. EEG data were recorded during resting state eyes-closed condition. Exact low-resolution brain electromagnetic tomography (eLORETA) estimated the power and functional connectivity of cortical sources in frontal, central, parietal, occipital, temporal, and limbic regions. Delta (2-4 Hz), theta (4-8 Hz), alpha 1 (8-10.5 Hz), alpha 2 (10.5-13 Hz), beta 1 (13-20 Hz), beta 2 (20-30 Hz), and gamma (30-40 Hz) were the frequency bands of interest. The classification rates of interest were those with an area under the receiver operating characteristic curve (AUROC) higher than 0.7 as a threshold for a moderate classification rate (i.e. 70%). Results showed that the following EEG markers overcame this threshold: (i) central, parietal, occipital, temporal, and limbic delta/alpha 1 current density; (ii) central, parietal, occipital temporal, and limbic delta/alpha 2 current density; (iii) frontal theta/alpha 1 current density; (iv) occipital delta/alpha 1 inter-hemispherical connectivity; (v) occipital-temporal theta/alpha 1 right and left intra-hemispherical connectivity; and (vi) parietal-limbic alpha 1 right intra-hemispherical connectivity. Occipital delta/alpha 1 current density showed the best classification rate (sensitivity of 73.3%, specificity of 78%, accuracy of 75.5%, and AUROC of 82%). These results suggest that EEG source markers can classify Nold and AD individuals with a moderate classification rate higher than 80%.

Background: Assessment and rating of Parkinson’s Disease (PD) are commonly based on the medical observation of several clinical manifestations, including the analysis of motor activities. In particular, medical specialists refer to the MDS-UPDRS (Movement Disorder Society – sponsored revision of Unified Parkinson’s Disease Rating Scale) that is the most widely used clinical scale for PD rating. However, clinical scales rely on the observation of some subtle motor phenomena that are either difficult to capture with human eyes or could be misclassified. This limitation motivated several researchers to develop intelligent systems based on machine learning algorithms able to automatically recognize the PD. Nevertheless, most of the previous studies investigated the classification between healthy subjects and PD patients without considering the automatic rating of different levels of severity. Methods: In this context, we implemented a simple and low-cost clinical tool that can extract postural and kinematic features with the Microsoft Kinect v2 sensor in order to classify and rate PD. Thirty participants were enrolled for the purpose of the present study: sixteen PD patients rated according to MDS-UPDRS and fourteen healthy paired subjects. In order to investigate the motor abilities of the upper and lower body, we acquired and analyzed three main motor tasks: (1) gait, (2) finger tapping, and (3) foot tapping. After preliminary feature selection, different classifiers based on Support Vector Machine (SVM) and Artificial Neural Networks (ANN) were trained and evaluated for the best solution. Results: Concerning the gait analysis, results showed that the ANN classifier performed the best by reaching 89.4% of accuracy with only nine features in diagnosis PD and 95.0% of accuracy with only six features in rating PD severity. Regarding the finger and foot tapping analysis, results showed that an SVM using the extracted features was able to classify healthy subjects versus PD patients with great performances by reaching 87.1% of accuracy. The results of the classification between mild and moderate PD patients indicated that the foot tapping features were the most representative ones to discriminate (81.0% of accuracy). Conclusions: The results of this study have shown how a low-cost vision-based system can automatically detect subtle phenomena featuring the PD. Our findings suggest that the proposed tool can support medical specialists in the assessment and rating of PD patients in a real clinical scenario.