Risk Stratification for Cardiotoxicity in Breast Cancer Patients: Predicting Early Decline of LVEF After Treatment.

This study introduces AI-based models in prediction and risk assessment of early cardiac dysfunction in older breast cancer patients, as a side-effect of their cancer treatment. Using only features extracted during the baseline evaluation of each patient the proposed methodology could predict a decline in LVEF values in 4 different follow-up intervals during the first year after treatment initiation (i.e. months 3-12), with a mean accuracy of 66.67% and up to 73.55%. Selected baseline predictive factors were ranked according to their prevalence in the evaluation experiments, replicating the importance of various cardiac disorders at baseline, LVEF value and a higher age, which are all previously reported, while introducing Diabetes as an important risk factor.Clinical Relevance- Healthcare providers can better assess cardiovascular health status and risk of cardiotoxicity in the cancer treatment continuum. This will enable timely intervention and close monitoring on high risk patients while saving resources for low risk patients.

Stature and mitigation systems affect the risk of leg injury in vehicles attacked under the body by explosive devices.

A finite-element (FE) model, previously validated for underbody blast (UBB) loading, was used here to study the effect of stature and of mitigation systems on injury risk to the leg. A range of potential UBB loadings was simulated. The risk of injury to the leg was calculated when no protection was present, when a combat boot (Meindl Desert Fox) was worn, and when a floor mat (IMPAXXTM), which can be laid on the floor of a vehicle, was added. The risk of injury calculated indicates that the floor mat provided a statistically significant reduction in the risk of a major calcaneal injury for peak impact speeds below 17.5 m/s when compared with the scenarios in which the floor mat was not present. The risk of injury to the leg was also calculated for a shorter and a taller stature compared to that of the nominal, 50th percentile male anthropometry; shorter and taller statures were constructed by scaling the length of the tibia of the nominal stature. The results showed that there is a higher risk of leg injury associated with the short stature compared to the nominal and tall statures, whereas the leg-injury risk between nominal and tall statures was statistically similar. These findings provide evidence that the combat boot and the floor mat tested here have an attenuating effect, albeit limited to a range of possible UBB loads. The effect of stature on injury has implications on how vehicle design caters for all potential anthropometries and indeed gender, as women, on average, are shorter than men. The results from the computational simulations here complement laboratory and field experimental models of UBB, and so they contribute to the improvement of UBB safety technology and strategy.

Harmonization Strategies in Multicenter MRI-Based Radiomics.

Radiomics analysis is a powerful tool aiming to provide diagnostic and prognostic patient information directly from images that are decoded into handcrafted features, comprising descriptors of shape, size and textural patterns. Although radiomics is gaining momentum since it holds great promise for accelerating digital diagnostics, it is susceptible to bias and variation due to numerous inter-patient factors (e.g., patient age and gender) as well as inter-scanner ones (different protocol acquisition depending on the scanner center). A variety of image and feature based harmonization methods has been developed to compensate for these effects; however, to the best of our knowledge, none of these techniques has been established as the most effective in the analysis pipeline so far. To this end, this review provides an overview of the challenges in optimizing radiomics analysis, and a concise summary of the most relevant harmonization techniques, aiming to provide a thorough guide to the radiomics harmonization process.

Segmentation of left atrium using CT images and a deep learning model.

The left atrium (LA) is one of the cardiac cavities with the most complex anatomical structures. Its role in the clinical diagnosis and patient's management is critical, as it is responsible for the atrial fibrillation, a condition that promotes the thrombogenesis inside the left atrial appendage. The development of an automated approach for LA segmentation is a demanding task mainly due to its anatomical complexity and the variation of its shape among patients. In this study, we focus to develop an unbiased pipeline capable to segment the atrial cavity from CT images. For evaluation purposes state-of-the-art metrics were used to assess the segmentation results. Particularly, the results indicated the mean values of the dice score 80%, the hausdorff distance 11.78mm, the average surface distance 2.24mm and the rand error index 0.2.

Comparison of three U-Net family architectures for left ventricular myocardial wall automatic segmentation.

Left ventricular (LV) segmentation is an important process which can provide quantitative clinical measurements such as volume, wall thickness and ejection fraction. The development of an automatic LV segmentation procedure is a challenging and complicated task mainly due to the variation of the heart shape from patient to patient, especially for those with pathological and physiological changes. In this study, we focus on the implementation, evaluation and comparison of three different Deep Learning architectures of the U-Net family: the custom 2-D U-Net, the ResU-Net++ and the DenseU-Net, in order to segment the LV myocardial wall. Our approach was applied to cardiac CT datasets specifically derived from patients with hypertrophic cardiomyopathy. The results of the models demonstrated high performance in the segmentation process with minor losses. The model revealed a dice score for U-Net, Res-U-net++ and Dense U-Net, 0.81, 0.82 and 0.84, respectively.

An Experimentally Validated Finite Element Model of the Lower Limb to Investigate the Efficacy of Blast Mitigation Systems.

Improvised explosive devices (IEDs) used in the battlefield cause damage to vehicles and their occupants. The injury burden to the casualties is significant. The biofidelity and practicality of current methods for assessing current protection to reduce the injury severity is limited. In this study, a finite-element (FE) model of the leg was developed and validated in relevant blast-loading conditions, and then used to quantify the level of protection offered by a combat boot. An FE model of the leg of a 35 years old male cadaver was developed. The cadaveric leg was tested physically in a seated posture using a traumatic injury simulator and the results used to calibrate the FE model. The calibrated model predicted hindfoot forces that were in good correlation (using the CORrelation and Analysis or CORA tool) with data from force sensors; the average correlation and analysis rating (according to ISO18571) was 0.842. The boundary conditions of the FE model were then changed to replicate pendulum tests conducted in previous studies which impacted the leg at velocities between 4 and 6.7 m/s. The FE model results of foot compression and peak force at the proximal tibia were within the experimental corridors reported in the studies. A combat boot was then incorporated into the validated computational model. Simulations were run across a range of blast-related loading conditions. The predicted proximal tibia forces and associated risk of injury indicated that the combat boot reduced the injury severity for low severity loading cases with higher times to peak velocity. The reduction in injury risk varied between 6 and 37% for calcaneal minor injuries, and 1 and 54% for calcaneal major injuries. No injury-risk reduction was found for high severity loading cases. The validated FE model of the leg developed here was able to quantify the protection offered by a combat boot to vehicle occupants across a range of blast-related loading conditions. It can now be used as a design and as an assessment tool to quantify the level of blast protection offered by other mitigation technologies.

A computational pipeline for data augmentation towards the improvement of disease classification and risk stratification models: A case study in two clinical domains.

Virtual population generation is an emerging field in data science with numerous applications in healthcare towards the augmentation of clinical research databases with significant lack of population size. However, the impact of data augmentation on the development of AI (artificial intelligence) models to address clinical unmet needs has not yet been investigated. In this work, we assess whether the aggregation of real with virtual patient data can improve the performance of the existing risk stratification and disease classification models in two rare clinical domains, namely the primary Sjögren's Syndrome (pSS) and the hypertrophic cardiomyopathy (HCM), for the first time in the literature. To do so, multivariate approaches, such as, the multivariate normal distribution (MVND), and straightforward ones, such as, the Bayesian networks, the artificial neural networks (ANNs), and the tree ensembles are compared against their performance towards the generation of high-quality virtual data. Both boosting and bagging algorithms, such as, the Gradient boosting trees (XGBoost), the AdaBoost and the Random Forests (RFs) were trained on the augmented data to evaluate the performance improvement for lymphoma classification and HCM risk stratification. Our results revealed the favorable performance of the tree ensemble generators, in both domains, yielding virtual data with goodness-of-fit 0.021 and KL-divergence 0.029 in pSS and 0.029, 0.027 in HCM, respectively. The application of the XGBoost on the augmented data revealed an increase by 10.9% in accuracy, 10.7% in sensitivity, 11.5% in specificity for lymphoma classification and 16.1% in accuracy, 16.9% in sensitivity, 13.7% in specificity in HCM risk stratification.

Wall shear stress alterations at left atrium and left atrial appendage employing abnormal blood velocity profiles.

In this study, we developed and analyzed different patient-specific 3D anatomical models of the left atrium including left atrial Appendage, in order to investigate the local hemodynamics. Particularly, we focused on the left atrial appendage and its impact on thrombus formation due to wall shear stress alterations. A 3D semi-automated reconstruction approach was carried out to segment and reconstruct the left atrium from CT scans. Six different patients were studied applying their patient-specific clinical data. Three different velocity profiles simulated for each patient case, representing one normal and two abnormal conditions. Simulations varied significantly according to different appendage morphologies. Our scope is to describe the hemodynamic behavior at the left atrium and the left atrial appendage according to different blood velocities based on their anatomic variety (chicken wing 0.14 m/s, windsock 0.10, cactus 0.08, and cauliflower 0.04). Wall shear stress results were demonstrated and correlated with the velocities and the thrombus formation inside the appendage cavity.

Lower Limb Posture Affects the Mechanism of Injury in Under-Body Blast.

Over 80% of wounded Service Members sustain at least one extremity injury. The 'deck-slap' foot, a product of the vehicle's floor rising rapidly when attacked by a mine to injure the limb, has been a signature injury in recent conflicts. Given the frequency and severity of these combat-related extremity injuries, they require the greatest utilisation of resources for treatment, and have caused the greatest number of disabled soldiers during recent conflicts. Most research efforts focus on occupants seated with both tibia-to-femur and tibia-to-foot angles set at 90°; it is unknown whether results obtained from these tests are applicable when alternative seated postures are adopted. To investigate this, lower limbs from anthropometric testing devices (ATDs) and post mortem human subjects (PMHSs) were loaded in three different seated postures using an under-body blast injury simulator. Using metrics that are commonly used for assessing injury, such as the axial force and the revised tibia index, the lower limb of ATDs were found to be insensitive to posture variations while the injuries sustained by the PMHS lower limbs differed in type and severity between postures. This suggests that the mechanism of injury depends on the posture and that this cannot be captured by the current injury criteria. Therefore, great care should be taken when interpreting and extrapolating results, especially in vehicle qualification tests, when postures other than the 90°-90° are of interest.

Material properties of the heel fat pad across strain rates.

The complex structural and material behaviour of the human heel fat pad determines the transmission of plantar loading to the lower limb across a wide range of loading scenarios; from locomotion to injurious incidents. The aim of this study was to quantify the hyper-viscoelastic material properties of the human heel fat pad across strains and strain rates. An inverse finite element (FE) optimisation algorithm was developed and used, in conjunction with quasi-static and dynamic tests performed to five cadaveric heel specimens, to derive specimen-specific and mean hyper-viscoelastic material models able to predict accurately the response of the tissue at compressive loading of strain rates up to 150s-1. The mean behaviour was expressed by the quasi-linear viscoelastic (QLV) material formulation, combining the Yeoh material model (C10=0.1MPa, C30=7MPa, K=2GPa) and Prony׳s terms (A1=0.06, A2=0.77, A3=0.02 for τ1=1ms, τ2=10ms, τ3=10s). These new data help to understand better the functional anatomy and pathophysiology of the foot and ankle, develop biomimetic materials for tissue reconstruction, design of shoe, insole, and foot and ankle orthoses, and improve the predictive ability of computational models of the foot and ankle used to simulate daily activities or predict injuries at high rate injurious incidents such as road traffic accidents and underbody blast.