Feasibility of proton dosimetry overriding planning CT with daily CBCT elaborated through generative artificial intelligence tools.

Radiotherapy commonly utilizes cone beam computed tomography (CBCT) for patient positioning and treatment monitoring. CBCT is deemed to be secure for patients, making it suitable for the delivery of fractional doses. However, limitations such as a narrow field of view, beam hardening, scattered radiation artifacts, and variability in pixel intensity hinder the direct use of raw CBCT for dose recalculation during treatment. To address this issue, reliable correction techniques are necessary to remove artifacts and remap pixel intensity into Hounsfield Units (HU) values. This study proposes a deep-learning framework for calibrating CBCT images acquired with narrow field of view (FOV) systems and demonstrates its potential use in proton treatment planning updates. Cycle-consistent generative adversarial networks (cGAN) processes raw CBCT to reduce scatter and remap HU. Monte Carlo simulation is used to generate CBCT scans, enabling the possibility to focus solely on the algorithm's ability to reduce artifacts and cupping effects without considering intra-patient longitudinal variability and producing a fair comparison between planning CT (pCT) and calibrated CBCT dosimetry. To showcase the viability of the approach using real-world data, experiments were also conducted using real CBCT. Tests were performed on a publicly available dataset of 40 patients who received ablative radiation therapy for pancreatic cancer. The simulated CBCT calibration led to a difference in proton dosimetry of less than 2%, compared to the planning CT. The potential toxicity effect on the organs at risk decreased from about 50% (uncalibrated) up the 2% (calibrated). The gamma pass rate at 3%/2 mm produced an improvement of about 37% in replicating the prescribed dose before and after calibration (53.78% vs 90.26%). Real data also confirmed this with slightly inferior performances for the same criteria (65.36% vs 87.20%). These results may confirm that generative artificial intelligence brings the use of narrow FOV CBCT scans incrementally closer to clinical translation in proton therapy planning updates.

Mixed Reality and Artificial Intelligence: A Holistic Approach to Multimodal Visualization and Extended Interaction in Knee Osteotomy.

OBJECTIVE: Recent advancements in augmented reality led to planning and navigation systems for orthopedic surgery. However little is known about mixed reality (MR) in orthopedics. Furthermore, artificial intelligence (AI) has the potential to boost the capabilities of MR by enabling automation and personalization. The purpose of this work is to assess Holoknee prototype, based on AI and MR for multimodal data visualization and surgical planning in knee osteotomy, developed to run on the HoloLens 2 headset. METHODS: Two preclinical test sessions were performed with 11 participants (eight surgeons, two residents, and one medical student) executing three times six tasks, corresponding to a number of holographic data interactions and preoperative planning steps. At the end of each session, participants answered a questionnaire on user perception and usability. RESULTS: During the second trial, the participants were faster in all tasks than in the first one, while in the third one, the time of execution decreased only for two tasks ("Patient selection" and "Scrolling through radiograph") with respect to the second attempt, but without statistically significant difference (respectively [Formula: see text] = 0.14 and [Formula: see text] = 0.13, [Formula: see text]). All subjects strongly agreed that MR can be used effectively for surgical training, whereas 10 (90.9%) strongly agreed that it can be used effectively for preoperative planning. Six (54.5%) agreed and two of them (18.2%) strongly agreed that it can be used effectively for intraoperative guidance. DISCUSSION/CONCLUSION: In this work, we presented Holoknee, the first holistic application of AI and MR for surgical planning for knee osteotomy. It reported promising results on its potential translation to surgical training, preoperative planning, and surgical guidance. Clinical and Translational Impact Statement - Holoknee can be helpful to support surgeons in the preoperative planning of knee osteotomy. It has the potential to impact positively the training of the future generation of residents and aid surgeons in the intraoperative stage.

Combined Edge Loss UNet for Optimized Segmentation in Total Knee Arthroplasty Preoperative Planning.

Bone segmentation and 3D reconstruction are crucial for total knee arthroplasty (TKA) surgical planning with Personalized Surgical Instruments (PSIs). Traditional semi-automatic approaches are time-consuming and operator-dependent, although they provide reliable outcomes. Moreover, the recent expansion of artificial intelligence (AI) tools towards various medical domains is transforming modern healthcare. Accordingly, this study introduces an automated AI-based pipeline to replace the current operator-based tibia and femur 3D reconstruction procedure enhancing TKA preoperative planning. Leveraging an 822 CT image dataset, a novel patch-based method and an improved segmentation label generation algorithm were coupled to a Combined Edge Loss UNet (CEL-UNet), a novel CNN architecture featuring an additional decoding branch to boost the bone boundary segmentation. Root Mean Squared Errors and Hausdorff distances compared the predicted surfaces to the reference bones showing median and interquartile values of 0.26 (0.19-0.36) mm and 0.24 (0.18-0.32) mm, and of 1.06 (0.73-2.15) mm and 1.43 (0.82-2.86) mm for the tibia and femur, respectively, outperforming previous results of our group, state-of-the-art, and UNet models. A feasibility analysis for a PSI-based surgical plan revealed sub-millimetric distance errors and sub-angular alignment uncertainties in the PSI contact areas and the two cutting planes. Finally, operational environment testing underscored the pipeline's efficiency. More than half of the processed cases complied with the PSI prototyping requirements, reducing the overall time from 35 min to 13.1 s, while the remaining ones underwent a manual refinement step to achieve such PSI requirements, performing the procedure four to eleven times faster than the manufacturer standards. To conclude, this research advocates the need for real-world applicability and optimization of AI solutions in orthopedic surgical practice.

Breathing motion pattern in cyclists: role of inferior against superior thorax compartment.

Thoracoabdominal breathing motion pattern is being considered in sports training because of its contribution, along with other physiological adaptations, to overall performance. We examined whether and how experience with cycling training modifies the thoracoabdominal motion patterns. We utilized optoelectronic plethysmography to monitor ten trained male cyclists and compared them to ten physically active male participants performing breathing maneuvers. Cyclists then participated in a self-paced time trial to explore the similarity between that observed during resting breathing. From the 3D coordinates of 32 markers positioned on each participant's trunk, we calculated the percentage of contribution of the superior thorax, inferior thorax, and abdomen and the correlation coefficient among these compartments. During the rest maneuvers, the cyclists showed a thoracoabdominal motion pattern characterized by an increased role of the inferior thorax relative to the superior thorax (26.69±5.88%, 34.93±5.03%; p = 0.002, respectively), in contrast to the control group (26.69±5.88%; 25.71±6.04%, p = 0.4, respectively). In addition, the inferior thorax showed higher coordination in phase with the abdomen. Furthermore, the results of the time trial test underscored the same pattern found in cyclists breathing at rest, suggesting that the development of a permanent modification in respiratory mechanics may be associated with cycling practice.

A new system for in-room image guidance in particle therapy at CNAO.

This paper describes the design, installation, and commissioning of an in-room imaging device developed at the Centro Nazionale di Adroterapia Oncologica (CNAO, Pavia, Italy). The system is an upgraded version of the one previously installed in 2014, and its design accounted for the experience gained in a decade of clinical practice of patient setup verification and correction through robotic-supported, off-isocenter in-room image guidance. The system's basic feature consists of image-based setup correction through 2D/3D and 3D/3D registration through a dedicated HW/SW platform. The major update with respect to the device already under clinical usage resides in the implementation of a functionality for extending the field of view of the reconstructed Cone Beam CT (CBCT) volume, along with improved overall safety and functional optimization. We report here details on the procedures implemented for system calibration under all imaging modalities and the results of the technical and preclinical commissioning of the device performed on two different phantoms. In the technical commissioning, specific attention was given to the assessment of the accuracy with which the six-degrees-of-freedom correction vector computed at the off-isocenter imaging position was propagated to the planned isocentric irradiation geometry. During the preclinical commissioning, the entire clinical-like procedure for detecting and correcting imposed, known setup deviation was tested on an anthropomorphic radioequivalent phantom. Results showed system performance within the sub-millimeter and sub-degree range according to project specifications under each imaging modality, making it ready for clinical application.

SLEEP-SEE-THROUGH: Explainable Deep Learning for Sleep Event Detection and Quantification From Wearable Somnography.

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.

Analytical Challenges in Diabetes Management: Towards Glycated Albumin Point-of-Care Detection.

Diabetes mellitus is a worldwide-spread chronic metabolic disease that occurs when the pancreas fails to produce enough insulin levels or when the body fails to effectively use the secreted pancreatic insulin, eventually resulting in hyperglycemia. Systematic glycemic control is the only procedure at our disposal to prevent diabetes long-term complications such as cardiovascular disorders, kidney diseases, nephropathy, neuropathy, and retinopathy. Glycated albumin (GA) has recently gained more and more attention as a control biomarker thanks to its shorter lifespan and wider reliability compared to glycated hemoglobin (HbA1c), currently the "gold standard" for diabetes screening and monitoring in clinics. Various techniques such as ion exchange, liquid or affinity-based chromatography and immunoassay can be employed to accurately measure GA levels in serum samples; nevertheless, due to the cost of the lab equipment and complexity of the procedures, these methods are not commonly available at clinical sites and are not suitable to home monitoring. The present review describes the most up-to-date advances in the field of glycemic control biomarkers, exploring in particular the GA with a special focus on the recent experimental analysis techniques, using enzymatic and affinity methods. Finally, analysis steps and fundamental reading technologies are integrated into a processing pipeline, paving the way for future point-of-care testing (POCT). In this view, we highlight how this setup might be employed outside a laboratory environment to reduce the time from measurement to clinical decision, and to provide diabetic patients with a brand-new set of tools for glycemic self-monitoring.

CEL-Unet: a novel CNN architecture for 3D Segmentation of Knee Bones affected by Severe Osteoarthritis for PSI-Based Surgical Planning.

Unet architectures are promising deep learning networks exploited to perform the automatic segmentation of bone CT images, in line with their ability to deal with pathological deformations and size-varying anatomies. However, bone degeneration, like the development of irregular osteophytes as well as mineral density alterations might interfere with this automated process and demand extensive manual refinement. The aim of this work is to implement an innovative Unet variant, the CEL-Unet, to improve the femur and tibia segmentation outcomes in osteoarthritic knee joints. In this network the decoding path is split into a region and contour-aware branch to increase the prediction reliability in such pathological conditions. The comparison between the segmentation results achieved with a standard Unet and its novel variant (CEL-Unet) was performed as follows: the Unet was trained with 5 different loss functions: Dice Loss, Focal Loss, Exponential Logarithmic Loss, Double Cross Entropy Loss and Distanced Cross Entropy loss. The CEL-Unet was instead trained with two loss functions, one for each of the network outputs, namely Mask and Edge, yielding the so-called Combined Edge Loss (CEL) function. A set of 259 knee CT scans was used to train the model and test segmentation performance. The CEL-Unet outperformed all other Unet-based models, reaching the highest Jaccard values of about 0.97 and 0.96 on femur and tibia, respectively. Clinical Relevance- With the increasing rate of Total Knee Arthoplasty deep learning-based methods can achieve fast accurate and automatic 3D segmentation of the knee joint bones to enhance new costumized pre-operative planning.

Identification of Characteristic Points in Multivariate Physiological Signals by Sensor Fusion and Multi-Task Deep Networks.

Identification of characteristic points in physiological signals, such as the peak of the R wave in the electrocardiogram and the peak of the systolic wave of the photopletismogram, is a fundamental step for the quantification of clinical parameters, such as the pulse transit time. In this work, we presented a novel neural architecture, called eMTUnet, to automate point identification in multivariate signals acquired with a chest-worn device. The eMTUnet consists of a single deep network capable of performing three tasks simultaneously: (i) localization in time of characteristic points (labeling task), (ii) evaluation of the quality of signals (classification task); (iii) estimation of the reliability of classification (reliability task). Preliminary results in overnight monitoring showcased the ability to detect characteristic points in the four signals with a recall index of about 1.00, 0.90, 0.90, and 0.80, respectively. The accuracy of the signal quality classification was about 0.90, on average over four different classes. The average confidence of the correctly classified signals, against the misclassifications, was 0.93 vs. 0.52, proving the worthiness of the confidence index, which may better qualify the point identification. From the achieved outcomes, we point out that high-quality segmentation and classification are both ensured, which brings the use of a multi-modal framework, composed of wearable sensors and artificial intelligence, incrementally closer to clinical translation.

Hybrid Convolutional Networks for End-to-End Event Detection in Concurrent PPG and PCG Signals Affected by Motion Artifacts.

The accurate detection of physiologically-related events in photopletismographic (PPG) and phonocardiographic (PCG) signals, recorded by wearable sensors, is mandatory to perform the estimation of relevant cardiovascular parameters like the heart rate and the blood pressure. However, the measurement performed in uncontrolled conditions without clinical supervision leaves the detection quality particularly susceptible to noise and motion artifacts. This work proposes a new fully-automatic computational framework, based on convolutional networks, to identify and localize fiducial points in time as the foot, maximum slope and peak in PPG signal and the S1 sound in the PCG signal, both acquired by a custom chest sensor, described recently in the literature by our group. The event detection problem was reframed as a single hybrid regression-classification problem entailing a custom neural architecture to process sequentially the PPG and PCG signals. Tests were performed analysing four different acquisition conditions (rest, cycling, rest recovery and walking). Cross-validation results for the three PPG fiducial points showed identification accuracy greater than 93 % and localization error (RMSE) less than 10 ms. As expected, cycling and walking conditions provided worse results than rest and recovery, however reaching an accuracy greater than 90 % and a localization error less than 15 ms. Likewise, the identification and localization error for S1 sound were greater than 90 % and less than 25 ms. Overall, this study showcased the ability of the proposed technique to detect events with high accuracy not only for steady acquisitions but also during subject movements. We also showed that the proposed network outperformed traditional Shannon-energy-envelope method in the detection of S1 sound, reaching detection performance comparable to state of the art algorithms. Therefore, we argue that coupling chest sensors and deep learning processing techniques may disclose wearable devices to unobtrusively acquire health information, being less affected by noise and motion artifacts.

Septic arthritis following joint injections: a 17 years retrospective study in an Academic General Hospital.

BACKGROUND: Septic arthritis following intra-articular infiltrations is an uncommon devastating complication correlated  to high costs for the health service and often to poor outcomes. The purpose of this study is to assess a 17-years experience in a single academic multispecialist hospital managing this uncommon complication in Orthopaedic practice. METHODS: Patients with diagnosis of septic arthritis following joint injections treated in our hospital from January 2002 to December 2019 were included in the study. Clinical and demographic data, pathogens, injected agent, conservative/surgical treatments were reviewed. Patient were classified according to the ore operative Charlson Comorbidity Index (CCI) and the Cierny-Mader Classification(CMC). Furthermore follow-up outcome and time occurred to infection eradication were registered. RESULTS: We included in the study 11 patients with a median age of 74 years old (IQR= 61.5 - 79). The median CCI was 3  (IQR= 2 - 5) and the majority of patients belong to CMC = B class. Septic arthritis occurred mainly following corticosteroids injections and more frequently involving knees. The pathogen more often isolated was Staphylococcus aureus. Five (45%) patients referred an history of multiple intrarticular injections. 7 patients (64%) had a complete resolution following an arthroscopic debridement, 4 (36%) patients underwent to a 2-stage replacement and one of them hesitated in an arthrodesis because of a recurrent periprothesic joint infection and extensor apparatus insufficiency. CONCLUSION: The authors observed a potential increased risk of septic arthritis following joint injection in patients with history of multiple injections and poor health/immunological conditions. They recommend an early arthroscopic debridement as the treatment of choice especially in septic knees  performed in a multispecialist dedicated center.

Image-based shading correction for narrow-FOV truncated pelvic CBCT with deep convolutional neural networks and transfer learning.

PURPOSE: Cone beam computed tomography (CBCT) is a standard solution for in-room image guidance for radiation therapy. It is used to evaluate and compensate for anatomopathological changes between the dose delivery plan and the fraction delivery day. CBCT is a fast and versatile solution, but it suffers from drawbacks like low contrast and requires proper calibration to derive density values. Although these limitations are even more prominent with in-room customized CBCT systems, strategies based on deep learning have shown potential in improving image quality. As such, this article presents a method based on a convolutional neural network and a novel two-step supervised training based on the transfer learning paradigm for shading correction in CBCT volumes with narrow field of view (FOV) acquired with an ad hoc in-room system. METHODS: We designed a U-Net convolutional neural network, trained on axial slices of corresponding CT/CBCT couples. To improve the generalization capability of the network, we exploited two-stage learning using two distinct data sets. At first, the network weights were trained using synthetic CBCT scans generated from a public data set, and then only the deepest layers of the network were trained again with real-world clinical data to fine-tune the weights. Synthetic data were generated according to real data acquisition parameters. The network takes a single grayscale volume as input and outputs the same volume with corrected shading and improved HU values. RESULTS: Evaluation was carried out with a leave-one-out cross-validation, computed on 18 unique CT/CBCT pairs from six different patients from a real-world dataset. Comparing original CBCT to CT and improved CBCT to CT, we obtained an average improvement of 6 dB on peak signal-to-noise ratio (PSNR), +2% on structural similarity index measure (SSIM). The median interquartile range (IQR) Hounsfield unit (HU) difference between CBCT and CT improved from 161.37 (162.54) HU to 49.41 (66.70) HU. Region of interest (ROI)-based HU difference was narrowed by 75% in the spongy bone (femoral head), 89% in the bladder, 85% for fat, and 83% for muscle. The improvement in contrast-to-noise ratio for these ROIs was about 67%. CONCLUSIONS: We demonstrated that shading correction obtaining CT-compatible data from narrow-FOV CBCTs acquired with a customized in-room system is possible. Moreover, the transfer learning approach proved particularly beneficial for such a shading correction approach.

Comparison of Supervised and Unsupervised Approaches for the Generation of Synthetic CT from Cone-Beam CT.

Due to major artifacts and uncalibrated Hounsfield units (HU), cone-beam computed tomography (CBCT) cannot be used readily for diagnostics and therapy planning purposes. This study addresses image-to-image translation by convolutional neural networks (CNNs) to convert CBCT to CT-like scans, comparing supervised to unsupervised training techniques, exploiting a pelvic CT/CBCT publicly available dataset. Interestingly, quantitative results were in favor of supervised against unsupervised approach showing improvements in the HU accuracy (62% vs. 50%), structural similarity index (2.5% vs. 1.1%) and peak signal-to-noise ratio (15% vs. 8%). Qualitative results conversely showcased higher anatomical artifacts in the synthetic CBCT generated by the supervised techniques. This was motivated by the higher sensitivity of the supervised training technique to the pixel-wise correspondence contained in the loss function. The unsupervised technique does not require correspondence and mitigates this drawback as it combines adversarial, cycle consistency, and identity loss functions. Overall, two main impacts qualify the paper: (a) the feasibility of CNN to generate accurate synthetic CT from CBCT images, which is fast and easy to use compared to traditional techniques applied in clinics; (b) the proposal of guidelines to drive the selection of the better training technique, which can be shifted to more general image-to-image translation.

Is age rating enough to investigate changes in breathing motion pattern associated with aging of physically active women?

The most common way to analyze the effect of aging on breathing is to divide subjects into age groups. However, in addition to the fact that there is no consensus in the literature regarding age group division, such design critically influences the interpretation of the effects attributed to aging. Thus, this study aimed to investigate the feasibility to distinguish different age groups from the 3D kinematic variables of breathing motion (i.e., markers' coordinate as a function of time allowing the calculation of compartmental volume variations) and to analyze whether the aging could influence these variables. Seventy-three physically active women aged 19-80 years performed quiet breathing and vital capacity maneuvers. To record the thoracoabdominal breathing motion, the 3D coordinates of 32 retroreflective markers positioned on the trunk were used to estimate the volume variation of the superior thorax, inferior thorax, and abdomen. The percentage of contribution and the correlation coefficient were calculated to analyze the breathing motion pattern from the estimated volumes. The k-means cluster analysis was performed to analyze the age group classification. Linear regression was performed to investigate whether age can predict changes in the breathing motion pattern. The results showed that physically active women could not be classified into age groups from breathing motion. Despite significant p values of the linear regression, the high variability of the data suggested that age itself is not enough to predict the changes in breathing motion pattern when non-sedentary women are considered.

Deep Learning for Automated Analysis of Cellular and Extracellular Components of the Foreign Body Response in Multiphoton Microscopy Images.

The Foreign body response (FBR) is a major unresolved challenge that compromises medical implant integration and function by inflammation and fibrotic encapsulation. Mice implanted with polymeric scaffolds coupled to intravital non-linear multiphoton microscopy acquisition enable multiparametric, longitudinal investigation of the FBR evolution and interference strategies. However, follow-up analyses based on visual localization and manual segmentation are extremely time-consuming, subject to human error, and do not allow for automated parameter extraction. We developed an integrated computational pipeline based on an innovative and versatile variant of the U-Net neural network to segment and quantify cellular and extracellular structures of interest, which is maintained across different objectives without impairing accuracy. This software for automatically detecting the elements of the FBR shows promise to unravel the complexity of this pathophysiological process.

Humeral head necrosis associated to shaft non-union with massive bone loss: a case report.

Humeral non-union is a rare complication in shaft fractures, as well as humeral head necrosis is a possible complication in fracture involving the proximal third especially in four-part fractures. The presence of head osteonecrosis and diaphyseal non-union in the same arm represents a formidable challenge for an orthopaedic surgeon. We could not find any similar report in the literature dealing with this issue thus far. We present a case of a 65 years old woman referred to our hospital being affected by an atrophic humeral diaphyseal non-union with a massive bone loss (>10cm) associated to a humeral head osteonecrosis following a previous surgical procedures with a clear loosening of the hardware. At our institution,she was treated with hardware removal and insertion of a diaphyseal antibiotic spacer with Gentamycin for 2 months suspecting an active septic process at the union site despite negative cultural exams. Finally, she was treated with a cemented modular humeral megaprosthesis. At 20 months follow up, the patient, despite a reduced shoulder range of motion, referred to a pain-free recovery to an almost normal lifestyle, including car driving with no major disturbances. This case suggests that, in extreme selected cases following several failed treatments, megaprosthesis can represent a viable solution, especially in huge bone loss associated to joint degeneration, to ensure an acceptable return to a normal lifestyle.

Predicting Knee Joint Instability Using a Tibio-Femoral Statistical Shape Model.

Statistical shape models (SSMs) are a well established computational technique to represent the morphological variability spread in a set of matching surfaces by means of compact descriptive quantities, traditionally called "modes of variation" (MoVs). SSMs of bony surfaces have been proposed in biomechanics and orthopedic clinics to investigate the relation between bone shape and joint biomechanics. In this work, an SSM of the tibio-femoral joint has been developed to elucidate the relation between MoVs and bone angular deformities causing knee instability. The SSM was built using 99 bony shapes (distal femur and proximal tibia surfaces obtained from segmented CT scans) of osteoarthritic patients. Hip-knee-ankle (HKA) angle, femoral varus-valgus (FVV) angle, internal-external femoral rotation (IER), tibial varus-valgus (TVV) angles, and tibial slope (TS) were available across the patient set. Discriminant analysis (DA) and logistic regression (LR) classifiers were adopted to underline specific MoVs accounting for knee instability. First, it was found that thirty-four MoVs were enough to describe 95% of the shape variability in the dataset. The most relevant MoVs were the one encoding the height of the femoral and tibial shafts (MoV #2) and the one representing variations of the axial section of the femoral shaft and its bending in the frontal plane (MoV #5). Second, using quadratic DA, the sensitivity results of the classification were very accurate, being all >0.85 (HKA: 0.96, FVV: 0.99, IER: 0.88, TVV: 1, TS: 0.87). The results of the LR classifier were mostly in agreement with DA, confirming statistical significance for MoV #2 (p = 0.02) in correspondence to IER and MoV #5 in correspondence to HKA (p = 0.0001), FVV (p = 0.001), and TS (p = 0.02). We can argue that the SSM successfully identified specific MoVs encoding ranges of alignment variability between distal femur and proximal tibia. This discloses the opportunity to use the SSM to predict potential misalignment in the knee for a new patient by processing the bone shapes, removing the need for measuring clinical landmarks as the rotation centers and mechanical axes.

A multi-metric registration strategy for the alignment of longitudinal brain images in pediatric oncology.

Survival of pediatric patients with brain tumor has increased over the past 20 years, and increasing evidence of iatrogenic toxicities has been reported. In follow-ups, images are acquired at different time points where substantial changes of brain morphology occur, due to childhood physiological development and treatment effects. To address the image registration complexity, we propose two multi-metric approaches (Mplus, Mdot), combining mutual information (MI) and normalized gradient field filter (NGF). The registration performance of the proposed metrics was assessed on a simulated dataset (Brainweb) and compared with those obtained by MI and NGF separately, using mean magnitude and mean angular errors. The most promising metric (Mplus) was then selected and tested on a retrospective dataset comprising 45 pediatric patients who underwent focal radiotherapy for brain cancer. The quality of the realignment was scored by a radiation oncologist using a perceived misalignment metric (PM). All patients but one were assessed as PM ≤ 2 (good alignment), but the remaining one, severely affected by hydrocephalus and pneumocephalus at the first MRI acquisition, scored PM = 5 (unacceptable). These preliminary findings suggest that Mplus might improve the registration accuracy in complex applications such as pediatric oncology, when data are acquired throughout the years of follow-up, and is worth investigating. Graphical abstract Graphical abstract showing the clinical workflow of the overall registration procedure including the three rigid steps, the fourth deformable step, the reference MRI and the registered MRI as well as the contoured ROIs. The registration performance is assessed by means of the Perceived Misalignment score (PM).

Moving system with action sport cameras: 3D kinematics of the walking and running in a large volume.

Traditionally, motion analysis in clinical laboratories using optoelectronic systems (MOCAP) is performed in acquisition volumes of limited size. Given the complexity and cost of MOCAP in larger volumes, action sports cameras (ASC) represent an alternative approach in which the cameras move along with the subject during the movement task. Thus, this study aims to compare ASC against a traditional MOCAP in the perspective of reconstructing walking and running movements in large spatial volumes, which extend over the common laboratory setup. The two systems, consisting of four cameras each, were closely mounted on a custom carrying structure endowed with wheels. Two different acquisition setups, namely steady and moving conditions, were taken into account. A devoted calibration procedure, using the same protocol for the two systems, enabled the reconstruction of surface markers, placed on voluntary subjects, during the two acquisition setups. The comparison was quantitatively expressed in terms of three-dimensional (3D) marker reconstruction and kinematic computation quality. The quality of the marker reconstruction quality was quantified by means of the mean absolute error (MAE) of inter-marker distance and two-stick angle. The kinematic computation quality was quantified by means of the measure of the knee angle reconstruction during walking and running trials. In order to evaluate the camera system and moving camera effects, we used a Wilcoxon rank sum test and a Kruskal Wallis test (post-hoc Tukey), respectively. The Spearman correlation coefficient (ρ) and the Wilcoxon rank sum test were applied to compare the kinematic data obtained by the two camera systems. We found small ASC MAE values (< 2.6mm and 1.3°), but they were significantly bigger than the MOCAP (< 0.7mm and 0.6°). However, for the human movement no significant differences were found between kinematic variables in walking and running acquisitions (p>0.05), and the motion patterns of the right-left knee angles between both systems were very similar (ρ>0.90, p<0.05). These results highlighted the promising results of a system that uses ASC based on the procedure of mobile cameras to follow the movement of the subject, allowing a less constrained movement in the direction in which the structure moves, compared to the traditional laboratory setup.

Representative 3D shape of the distal femur, modes of variation and relationship with abnormality of the trochlear region.

The anatomy of the distal femur has a predominant influence on the mechanics of both patello- and tibio-femoral joints. Especially, the morphological degeneration of the trochlear region dramatically affects the overall knee biomechanics and, from a clinical point of view, the staging of such a degeneration is fundamental to tailor the optimal therapeutic solution. The description of morphological variability and pathological inter-subject differences of the trochlea can be achieved by means of statistical shape modeling of a set of three-dimensional surfaces. This representation encodes information, spread into the dataset, in terms of modes of variations that model global, regional and even local morphological features. In view of that, the aim of this study was to develop a statistical shape model of the distal femur to capture the variability of the trochlear region into specific modes of variation and to study the interplay between the variation of the trochlea and the condylar regions. Using CT scans of patients affected by different levels of abnormality of the trochlear region, the distal femur geometries were co-registered to a reference shape using the pair-wise correspondence approach and principal component analysis provided the key modes of variation (MoVs). Apart from the first two MoVs, which described the global magnitude of the femur and the shaft length, the main following ones showed high correlation with sulcus depth (r2=0.70), sulcus angle (r2=0.70), lateral trochlear inclination (r2=0.66), and height of the two condylar facets in the anterior direction (r2=0.66), whose abnormal variations are typical signs of trochlear degeneration. High interplay between trochlear abnormalities and notch width (r2=0.71), lateral condylar size (r2=0.67), and medial condylar size (r2=0.99) was found. Interestingly, the model predicted morphological associations not included in the training dataset, nonetheless difficult to demonstrate physiologically. Interestingly from a biomechanical point of view, the distribution of some MoVs was found statistically different across the patients featuring physiological and pathological ranges of hip-knee-ankle alignment, femoral internal-external rotation and tibial slope. However, no linear correlation was found between the angular indexes and such MoVs. As a result, we can assert that statistical modeling of the distal femur are to date an effective way to visualize and quantify abnormalities of the trochlear regions supporting the introduction of advanced analysis, diagnostic and treatment support tools to elucidate physiologic and pathological variability in the morphology, to drive the staging and assist the selection of the optimal treatment option tailored to the patient.