Osteoarthritis is a degenerative condition that affects bones and cartilage, often leading to structural changes, including osteophyte formation, bone density loss, and the narrowing of joint spaces. Over time, this process may disrupt the glenohumeral (GH) joint functionality, requiring a targeted treatment. Various options are available to restore joint functions, ranging from conservative management to surgical interventions, depending on the severity of the condition. This work introduces an innovative deep learning framework to process shoulder CT scans. It features the semantic segmentation of the proximal humerus and scapula, the 3D reconstruction of bone surfaces, the identification of the GH joint region, and the staging of three common osteoarthritic-related conditions: osteophyte formation (OS), GH space reduction (JS), and humeroscapular alignment (HSA). Each condition was stratified into multiple severity stages, offering a comprehensive analysis of shoulder bone structure pathology. The pipeline comprised two cascaded CNN architectures: 3D CEL-UNet for segmentation and 3D Arthro-Net for threefold classification. A retrospective dataset of 571 CT scans featuring patients with various degrees of GH osteoarthritic-related pathologies was used to train, validate, and test the pipeline. Root mean squared error and Hausdorff distance median values for 3D reconstruction were 0.22 mm and 1.48 mm for the humerus and 0.24 mm and 1.48 mm for the scapula, outperforming state-of-the-art architectures and making it potentially suitable for a PSI-based shoulder arthroplasty preoperative plan context. The classification accuracy for OS, JS, and HSA consistently reached around 90% across all three categories. The computational time for the entire inference pipeline was less than 15 s, showcasing the framework's efficiency and compatibility with orthopedic radiology practice. The achieved reconstruction and classification accuracy, combined with the rapid processing time, represent a promising advancement towards the medical translation of artificial intelligence tools. This progress aims to streamline the preoperative planning pipeline, delivering high-quality bone surfaces and supporting surgeons in selecting the most suitable surgical approach according to the unique patient joint conditions.

Evidence is rapidly accumulating that multifactorial nocturnal monitoring, through the coupling of wearable devices and deep learning, may be disruptive for early diagnosis and assessment of sleep disorders. In this work, optical, differential air-pressure and acceleration signals, acquired by a chest-worn sensor, are elaborated into five somnographic-like signals, which are then used to feed a deep network. This addresses a three-fold classification problem to predict the overall signal quality (normal, corrupted), three breathing-related patterns (normal, apnea, irregular) and three sleep-related patterns (normal, snoring, noise). In order to promote explainability, the developed architecture generates additional information in the form of qualitative (saliency maps) and quantitative (confidence indices) data, which helps to improve the interpretation of the predictions. Twenty healthy subjects enrolled in this study were monitored overnight for approximately ten hours during sleep. Somnographic-like signals were manually labeled according to the three class sets to build the training dataset. Both record- and subject-wise analyses were performed to evaluate the prediction performance and the coherence of the results. The network was accurate (0.96) in distinguishing normal from corrupted signals. Breathing patterns were predicted with higher accuracy (0.93) than sleep patterns (0.76). The prediction of irregular breathing was less accurate (0.88) than that of apnea (0.97). In the sleep pattern set, the distinction between snoring (0.73) and noise events (0.61) was less effective. The confidence index associated with the prediction allowed us to elucidate ambiguous predictions better. The saliency map analysis provided useful insights to relate predictions to the input signal content. While preliminary, this work supported the recent perspective on the use of deep learning to detect particular sleep events in multiple somnographic signals, thus representing a step towards bringing the use of AI-based tools for sleep disorder detection incrementally closer to clinical translation.

This paper describes a patient-specific mathematical model to predict the evolution of uterine cervical tumors at a macroscopic scale, during fractionated external radiotherapy. The model provides estimates of tumor regrowth and dead-cell reabsorption, incorporating the interplay between tumor regression rate and radiosensitivity, as a function of the tumor oxygenation level. Model parameters were estimated by minimizing the difference between predicted and measured tumor volumes, these latter being obtained from a set of 154 serial cone-beam computed tomography scans acquired on 16 patients along the course of the therapy. The model stratified patients according to two different estimated dynamics of dead-cell removal and to the predicted initial value of the tumor oxygenation. The comparison with a simpler model demonstrated an improvement in fitting properties of this approach (fitting error average value <5%, p < 0.01), especially in case of tumor late responses, which can hardly be handled by models entailing a constant radiosensitivity, failing to model changes from initial severe hypoxia to aerobic conditions during the treatment course. The model predictive capabilities suggest the need of clustering patients accounting for cancer cell line, tumor staging, as well as microenvironment conditions (e.g., oxygenation level).

Mathematical modeling of tumor response to radiotherapy has the potential of enhancing the quality of the treatment plan, which can be even tailored on an individual basis. Lack of extensive in vivo validation has prevented, however, reliable clinical translation of modeling outcomes. Image-guided radiotherapy is a consolidated treatment modality based on computed tomographic (CT) imaging for tumor delineation and volumetric cone beam CT data for periodic checks during treatment. In this study, a macroscopic model of tumor growth and radiation response is proposed, being able to adapt along the treatment course as volumetric tumor data become available. Model parameter learning was based on cone beam CT images in 13 uterine cervical cancer patients, subdivided into three groups (G1, G2, G3) according to tumor type and treatment. Three group-specific parameter sets (PS1, PS2, and PS3) on one general parameter set (PSa) were applied. The corresponding average model fitting errors were 14%, 18%, 13%, and 21%, respectively. The model adaptation testing was performed using volume data of three patients, other than the ones involved in the parameter learning. The extrapolation performance of the general model was improved, while comparable prediction errors were found for the group-specific approach. This suggests that an online parameter tuning can overcome the limitations of a suboptimal patient stratification, which appeared otherwise a critical issue.

Abstract Summary: GAAS, Gene Array Analyzer Software supports multi-user efficient management and suitable analyses of large amounts of gene expression data across replicated experiments. Its management framework handles input data generated by different technologies. A multi-user environment allows each user to store his/her own data visualization scheme, analysis parameters used, values and formats of the output data. The analysis engine performs: background and spot quality evaluation, data normalization, differential gene expression analyses in single and multiple replica experiments. Results of expression profiles can be interactively navigated through graphical interfaces and stored into output databases. Availability: http://www.medinfopoli.polimi.it/GAAS/ Contact: [email protected] Supplementary information: http://www.medinfopoli.polimi.it/GAAS/ * To whom correspondence should be addressed.

Calibration is a crucial step to obtaining 3D measurement using video camera-based stereo systems. All the parameters can be determined except for the pair of principal points, which poses a considerable drawback. Whereas in low-accuracy systems such points can be assumed to lie at the image center without degrading the overall 3D accuracy; in high-accuracy systems their true position must be computed accurately. In this case, all the calibration parameters can still be estimated through epipolar geometry, but it is necessary to minimize a highly nonlinear cost function. It is shown here that by combining two evolutionary optimization strategies this minimization can be carried out, both efficiently and reliably. The resulting strategy, which we call enhanced evolutionary search (EES), allows the full calibration of a stereo system using only a rigid bar. Moreover, EES can be applied to a wide range of applications where the cost function contains complex nonlinear relationships among the optimization variables.

One of the major problems which arises in the field of virtual design is the realization of virtual mannequins able to move in a human like way. This work focuses on the analysis of the human sitting working posture, which is described by a 30-DOF mannequin, modeling the upper part of the body (pelvis, trunk, arms, and head). Trajectories formation in point to point reaching movements represents the main topic. Our approach is based on the acquisition of real human kinematics data, collected by means of an automatic motion analyzer. Starting from the kinematics database of one subject, sit in front of a desk, a neural network was trained in order to generate the movements of the virtual mannequin. The work is divided into four parts: mannequin modeling, 3D human data collection, data preprocessing according to the biomechanical model, and design and training of a multilayer perceptron neural network.