Clustering blood donors via mixtures of product partition models with covariates.

Motivated by the problem of accurately predicting gap times between successive blood donations, we present here a general class of Bayesian nonparametric models for clustering. These models allow for the prediction of new recurrences, accommodating covariate information that describes the personal characteristics of the sample individuals. We introduce a prior for the random partition of the sample individuals, which encourages two individuals to be co-clustered if they have similar covariate values. Our prior generalizes product partition models with covariates (PPMx) models in the literature, which are defined in terms of cohesion and similarity functions. We assume cohesion functions that yield mixtures of PPMx models, while our similarity functions represent the denseness of a cluster. We show that including covariate information in the prior specification improves the posterior predictive performance and helps interpret the estimated clusters in terms of covariates in the blood donation application.

Reconsidering the Impact of Endovascular Repair on Short-Term and Mid-Term Outcomes in Peripheral Arterial Disease: A Retrospective Analysis.

BACKGROUND: The aim of the study is to compare the short-term and medium-term outcomes in patients who underwent open repair (OR) or endovascular repair (ER) for peripheral arterial disease (PAD) also including stratifications based on severity and year of the first intervention. METHODS: We conducted an observational retrospective single-center cohort study. We evaluated patients with PAD that primarily underwent ER, OR, minor, and major amputations in a single center from 2005 to 2020. The patients were then subdivided according to the type of intervention (OR versus ER), and stratified according to the International Classification of Diseases 9 code reported in the operating documents and to the year intervention. Mortality, minor, and major amputation rates occurring at 30 days, 2 years, and 5 years after the first intervention were evaluated as primary outcomes and compared between patient groups in both stratifications. Moreover, Kaplan-Maier curves were analyzed for these outcomes. RESULTS: One thousand four hundred ninety two patients (67.0% males) with PAD were evaluated. Their clinical presentations were intermittent claudication in 51.4% of cases, rest pain in 16.8%, ulcers in 10.3%, and gangrene in 21.5%. Nine hundred ninety seven (66.8%) underwent OR and 495 (33.2%) ER as first intervention for PAD. No statistical differences were observed in terms of mortality in the 2 groups (OR versus ER, P = 1,000, P = 0.357, and P = 0.688 at 30 days, 2 years, and 5 years, respectively). The rate of minor amputations was significantly higher (P < 0.012, P < 0.002, and P < 0.007 at 30 days, 2 years, and 5 years, respectively) for ER group in any of the observed follow-up periods. Also, we have observed that OR and ER do not have any significant short-term and medium-term major amputation rate differences. CONCLUSIONS: In our experience, the impact of ER does not significantly change short-term and mid-term major outcomes in patients with PAD.

Temporary Reperfusion of the Aneurysm Sac as a Prevention of Spinal Cord Ischemia After Endovascular Treatment of Thoracoabdominal Aortic Aneurysm: Systematic Review and Meta-analysis.

BACKGROUND: Spinal cord ischemia (SCI) is still a feared complication for patients suffering from thoracoabdominal aortic aneurysm (TAAA) who undergo endovascular treatment. The aims of this work are to review the available literature on different reperfusion methods of the aneurysm sac, and to analyze whether the different reperfusion methods, also in combination with other factors, are effective in reducing SCI risk and if the impact varies with the patient's age. METHODS: PubMed/MEDLINE library was searched for studies published until November 2020 concerning TAAA, endovascular repair, and SCI preventive measures. Systematic review and meta-analysis were conducted according to Preferred Reporting Items for Systematic reviews and Meta-Analyses criteria. Primary outcome consisted of correlation between endovascular repair techniques (type A: single step; type B: staged approach with reperfusion branches; type C: staged sequential approach with positioning of the thoracic component). A logistic-weighted regression for each event (SCI, transient, and permanent) was then performed with type of treatment, age, and interaction between them as input factors. Finally, another logistic-weighted regression was performed to analyze the other relevant factors for which observations are available together with the endovascular technique. RESULTS: Data from 53 studies with a total of 3095 patients were analyzed. Type A, type B, and type C endovascular strategies were adopted in 75%, 13%, and 12% of studied patients, respectively. Data showed that both type B and type C treatments are associated with lower risk of SCI, with a higher reduction of type C with respect to type B, although this positive trend is limited for elder patients. Moreover, a greater aortic diameter, a reduced aneurysm extent, and the absence of cerebrospinal fluid drainage positioning contribute to lower the risk of SCI. Concerning permanent SCI, both type B and type C are effective in reducing percentages for all ages, with type C treatment more beneficial for younger patients and type B for elder ones. CONCLUSION: According to the anatomy and the endovascular repair feasibility criteria, staged endovascular treatment appears to offer relevant advantages over single-step treatment in reducing the risk of SCI, regardless of the reperfusion method adopted.

Learning Curve Analysis of Robotic-Assisted Mitral Valve Repair with COVID-19 Exogenous Factor: A Single Center Experience.

Background and objective Renewed interest in robot-assisted cardiac procedures has been demonstrated by several studies. However, concerns have been raised about the need for a long and complex learning curve. In addition, the COVID-19 pandemic in 2020 might have affected the learning curve of these procedures. In this study, we investigated the impact of COVID-19 on the learning curve of robotic-assisted mitral valve surgery (RAMVS). The aim was to understand whether or not the benefits of RAMVS are compromised by its learning curve. Materials and Methods Between May 2019 and March 2023, 149 patients underwent RAMVS using the Da Vinci® X Surgical System at the Humanitas Gavazzeni Hospital, Bergamo, Italy. The selection of patients enrolled in the study was not influenced by case complexity. Regression models were used to formalize the learning curves, where preoperative data along with date of surgery and presence of COVID-19 were treated as the input covariates, while intraoperative and postoperative data were analyzed as output variables. Results The age of patients was 59.1 ± 13.3 years, and 70.5% were male. In total, 38.2% of the patients were operated on during the COVID-19 pandemic. The statistical analysis showed the positive impact of the learning curve on the trend of postoperative parameters, progressively reducing times and other key indicators. Focusing on the COVID-19 pandemic, statistical analysis did not recognize an impact on postoperative outcomes, although it became clear that variables not directly related to the intervention, especially ICU hours, were strongly influenced by hospital logistics during COVID-19. Conclusions Understanding the learning curve of robotic surgical procedures is essential to ensure their effectiveness and benefits. The learning curve involves not only surgeons but also other health care providers, and establishing a stable team in the early stage, as in our case, is important to shorten the duration. In fact, an exogenous factor such as the COVID-19 pandemic did not affect the robotic program despite the fact that the pandemic occurred early in the program.

Prediction model of isolated iliac and abdominal aneurysms.

OBJECTIVES: We analyse the cardiovascular risk factors in patients undergoing screening for Isolated Iliac Aneurysm (IIA) and Abdominal Aortic Aneurysm (AAA) and propose a logistic regression model to indicate patients at risk of IIA and/or AAA. METHODS: A screening programme was carried out to identify the presence of aneurysm based on Duplex scan examination. Cardiovascular risk factors information was collected from each subject. A descriptive analysis for the incidence of IIA and AAA stratified by age and sex was carried out to evaluate factors incidence. A logistic regression model was developed to predict the probability of developing an aneurysm based on the observed risk factor levels. A threshold probability of aneurysm risk for a datum patient was also identified to effectively direct screening protocols to patients most at risk. RESULTS: A cohort of 10 842 patients was evaluated: 1.52% affected by IIA, 2.69% by AAA and 3.90% by at least one. Risk factors analysis showed that: IIA was correlated with cardiological status, diabetes, cardiovascular disease family history, and dyslipidaemia; AAA was correlated with cardiological status, body mass index, hypertension, and dyslipidaemia; diabetes and dyslipidaemia were the most relevant factors with at least one aneurysm. The prediction tool based on the logistic regression and the threshold probability predict the presence of IIA and AAA in 69.7% and 83.8% of cases, under k-fold cross-validation. CONCLUSIONS: The proposed regression model can represent a valid aid to predict IIA and AAA presence and to select patients to be screened.

A discrete-event simulation model for analysing and improving operations in a blood donation centre.

BACKGROUND AND OBJECTIVES: Healthcare systems require effective and efficient blood donation supply chains to provide an adequate amount of whole blood and blood components to hospitals and transfusion centres. However, some crucial steps of the chain, for example blood collection, are not adequately studied in the literature. This work analyses the operations in a blood collection centre with the twofold aim of analysing different configurations and evaluating the effectiveness and feasibility of schedules defined at higher planning levels. MATERIALS AND METHODS: The analyses are performed through a discrete event simulation (DES) model that describes a customizable collection centre. Moreover, a feedback loop from the DES to the higher planning level allows to adjust scheduling decisions if they determine criticalities or infeasibilities at the operational level. RESULTS: Numerical tests have been conducted considering a real Italian provider. An experimental plan has been designed to compare different configurations for the blood collection centre and evaluate the best ones in terms of cost and service quality for the three main actors involved (donors, workers and managers). The best configurations have been also used to test the feedback loop. CONCLUSIONS: Results confirm the appropriateness of the proposed DES model, which can be considered a useful decision support tool for dimensioning and managing a blood collection centre, either as a standalone tool or in conjunction with a scheduler.

Home care vehicle routing problem with chargeable overtime and strict and soft preference matching.

A new scheduling problem arising in the home care context is addressed, whose novelty with respect to the literature lies in the way overtime is paid. In this problem, some clients are willing to pay a higher fee to cover the additional overtime cost, if such overtime is incurred because a caregiver works extra time with the client to preserve continuity of care. These overtime hours charged to clients unburden the company, which no longer has to balance between cost and continuity of care in a traditional way. The problem is also studied in a context that includes preferences expressed by both clients and caregivers. Strict preferences must be respected with a high priority, while soft preferences increase the satisfaction and should be preferably respected. We formalize the problem as a Mixed Integer Linear Problem and also propose a cluster-based decomposition to solve real-life instances. The problem is inspired by the real case study of a provider operating in the USA. Numerical results validate the model and confirm the capability of the decomposition approach to deal with real-life instances.

A novel bayesian approach with conditional autoregressive specification for intravoxel incoherent motion diffusion-weighted MRI.

The Intra-Voxel Incoherent Motion (IVIM) model is largely adopted to estimate slow and fast diffusion coefficients of water molecules in biological tissues, which are used in cancer applications. The most reported fitting approach is a voxel-wise segmented non-linear least square, whereas Bayesian approaches with a direct fit, also considering spatial regularization, were proposed too. In this work a novel segmented Bayesian method was proposed, also in combination with a spatial regularization through a Conditional Autoregressive (CAR) prior specification. The two segmented Bayesian approaches, with and without CAR specification, were compared with two standard least-square and a direct Bayesian fitting methods. All approaches were tested on simulated images and real data of patients with head-and-neck and rectal cancer. Estimation accuracy and maps noisiness were quantified on simulated images, whereas the coefficient of variation and the goodness of fit were evaluated for real data. Both versions of the segmented Bayesian approach outperformed the standard methods on simulated images for pseudo-diffusion (D∗ ) and perfusion fraction (f), whilst the segmented least-square fitting remained the less biased for the diffusion coefficient (D). On real data, Bayesian approaches provided the less noisy maps, and the two Bayesian methods without CAR generally estimated lower values for f and D∗ coefficients with respect to the other approaches. The proposed segmented Bayesian approaches were superior, in terms of estimation accuracy and maps quality, to the direct Bayesian model and the least-square fittings. The CAR method improved the estimation accuracy, especially for D∗ .

A stochastic risk-averse framework for blood donation appointment scheduling under uncertain donor arrivals.

Blood is a key resource in all health care systems, usually drawn from voluntary donors. We focus on the operations management in blood collection centers, which is a key step to guarantee an adequate blood supply and a good quality of service to donors, by addressing the so-called Blood Donation Appointment Scheduling problem. Its goal is to employ appointment scheduling to balance the production of blood units between days, in order to provide a reasonably constant supply to transfusion centers and hospitals, and reduce non-alignments between physicians' working times and donor arrivals at the collection center. We consider a two-phase solution framework taken from the literature, in which a deterministic linear programming model preallocates time slots to different blood types and a prioritization policy assigns the preallocated slots to the donors when they make a reservation. However, the problem is stochastic in nature and requires consideration of the uncertain arrivals of non-booked donors. In this work, to include the uncertain arrivals, we propose three stochastic counterparts of the preallocation model based on a risk-neutral objective and two risk-averse objectives, respectively, where the Conditional Value-at-Risk is considered as the risk measure in the last two methods. The resulting stochastic frameworks have been tested considering the historical data of one of the largest Italian collection centers, the Milan Department of the "Associazione Volontari Italiani Sangue" (AVIS). Results show the effectiveness of the stochastic models, especially the mean-risk one, and the need to include the uncertainty of arrivals in order to better balance the production of blood units.

Stent-Graft Deployment Increases Aortic Stiffness in an Ex Vivo Porcine Model.

BACKGROUND: Aortic stiffness is an independent predictor of cardiovascular mortality. In this study, the effect of thoracic endovascular aortic repair (TEVAR) on aortic stiffness is investigated by measuring aortic pulse wave velocity (PWV) in an ex vivo porcine model. METHODS: Fifteen fresh porcine thoracic aortas were connected to a benchtop pulsatile system. Intraluminal pressures were recorded in the ascending aorta and at the celiac trunk using a needle connected to a pressure sensor. The distance between the needles was divided by the time difference between the base of the pressure peaks to calculate aortic PWV at baseline and after stent-graft deployment and distal stent-graft extension. RESULTS: Mean aortic PWV was 5.0 m/s at baseline. PWV increased by 4% after proximal stent-graft deployment (P = 0.09) and by 18% after stent-graft extension (P < 0.001). Pulse pressure in the nonstented ascending aorta increased by 11.0 ± 1.2 mm Hg after proximal stent-graft deployment (P < 0.001) and by 17.3 ± 1.5 mm Hg after stent-graft extension (P < 0.001). The increases in PWV and pulse pressure showed a positive linear correlation with the percentage of stent-graft coverage (P < 0.001 and P < 0.001). CONCLUSIONS: In this experimental setup, aortic stiffness increased after stent-graft deployment, dependent on the percentage of the aorta that was covered by stent graft. These results show that TEVAR leads to significant changes in aortic hemodynamics, which merits evaluation in the clinical setting.

Assigning probabilities to qualitative dynamics of gene regulatory networks.

Mathematical and computational modeling frameworks play the leading role in the analysis and prediction of the dynamics of gene regulatory networks. The literature abounds in various approaches, all of which characterized by strengths and weaknesses. Among the others, Ordinary Differential Equations (ODE) models give a more general and detailed description of the network structure. But, analytical computations might be prohibitive as soon as the network dimension increases, and numerical simulation could be nontrivial, time-consuming and very often impracticable as precise and quantitative information on model parameters are frequently unknown and hard to estimate from experimental data. Last but not least, they do not account for the intrinsic stochasticity of regulation. In the present paper we consider a class of ODE models with stochastic parameters. The proposed method separates the deterministic aspects from the stochastic ones. Under specific assumptions and conditions, all possible trajectories of an ODE model, where incomplete knowledge of parameter values is symbolically and qualitatively expressed by initial inequalities only, are simulated in a single run from an initial state. Then, the occurrence probability of each trajectory, characterized by a sequence of parameter inequalities, is computed by associating probability density functions with network parameters. As demonstrated by its application to the gene repressilator system, the method seems particularly promising in the design of synthetic networks.

Uncertainty quantification in multi-class segmentation: Comparison between Bayesian and non-Bayesian approaches in a clinical perspective.

BACKGROUND: Automatic segmentation techniques based on Convolutional Neural Networks (CNNs) are widely adopted to automatically identify any structure of interest from a medical image, as they are not time consuming and not subject to high intra- and inter-operator variability. However, the adoption of these approaches in clinical practice is slowed down by some factors, such as the difficulty in providing an accurate quantification of their uncertainty. PURPOSE: This work aims to evaluate the uncertainty quantification provided by two Bayesian and two non-Bayesian approaches for a multi-class segmentation problem, and to compare the risk propensity among these approaches, considering CT images of patients affected by renal cancer (RC). METHODS: Four uncertainty quantification approaches were implemented in this work, based on a benchmark CNN currently employed in medical image segmentation: two Bayesian CNNs with different regularizations (Dropout and DropConnect), named BDR and BDC, an ensemble method (Ens) and a test-time augmentation (TTA) method. They were compared in terms of segmentation accuracy, using the Dice score, uncertainty quantification, using the ratio of correct-certain pixels (RCC) and incorrect-uncertain pixels (RIU), and with respect to inter-observer variability in manual segmentation. They were trained with the Kidney and Kidney Tumor Segmentation Challenge launched in 2021 (Kits21), for which multi-class segmentations of kidney, RC, and cyst on 300 CT volumes are available. Moreover, they were tested considering this and other two public renal CT datasets. RESULTS: Accuracy results achieved large differences across the structures of interest for all approaches, with an average Dice score of 0.92, 0.58, and 0.21 for kidney, tumor, and cyst, respectively. In terms of uncertainties, TTA provided the highest uncertainty, followed by Ens and BDC, whereas BDR provided the lowest, and minimized the number of incorrect certain pixels worse than the other approaches. Again, large differences were seen across the three structures in terms of RCC and RIU. These metrics were associated with different risk propensity, as BDR was the most risk-taking approach, able to provide higher accuracy in its prediction, but failing to assign uncertainty on incorrect segmentation in every case. The other three approaches were more conservative, providing large uncertainty regions, with the drawback of giving alert also on correct areas. Finally, the analysis of the inter-observer segmentation variability showed a significant variation among the four approaches on the external dataset, with BDR reporting the lowest agreement (Dice = 0.82), and TTA obtaining the highest score (Dice = 0.94). CONCLUSIONS: Our outcomes highlight the importance of quantifying the segmentation uncertainty and that decision-makers can choose the approach most in line with the risk propensity degree required by the application and their policy.

Radiomic Gradient in Peritumoural Tissue of Liver Metastases: A Biomarker for Clinical Practice? Analysing Density, Entropy, and Uniformity Variations with Distance from the Tumour.

The radiomic analysis of the tissue surrounding colorectal liver metastases (CRLM) enhances the prediction accuracy of pathology data and survival. We explored the variation of the textural features in the peritumoural tissue as the distance from CRLM increases. We considered patients with hypodense CRLMs >10 mm and high-quality computed tomography (CT). In the portal phase, we segmented (1) the tumour, (2) a series of concentric rims at a progressively increasing distance from CRLM (from one to ten millimetres), and (3) a cylinder of normal parenchyma (Liver-VOI). Sixty-three CRLMs in 51 patients were analysed. Median peritumoural HU values were similar to Liver-VOI, except for the first millimetre around the CRLM. Entropy progressively decreased (from 3.11 of CRLM to 2.54 of Liver-VOI), while uniformity increased (from 0.135 to 0.199, p < 0.001). At 10 mm from CRLM, entropy was similar to the Liver-VOI in 62% of cases and uniformity in 46%. In small CRLMs (≤30 mm) and responders to chemotherapy, normalisation of entropy and uniformity values occurred in a higher proportion of cases and at a shorter distance. The radiomic analysis of the parenchyma surrounding CRLMs unveiled a wide halo of progressively decreasing entropy and increasing uniformity despite a normal radiological aspect. Underlying pathology data should be investigated.

Inertance estimation in a lumped-parameter hydraulic simulator of human circulation.

Pulsatile mock loop systems are largely used to investigate the cardiovascular system in vitro. They consist of a pump, which replicates the heart, coupled with a lumped-parameter hydraulic afterload, which simulates vasculature. An accurate dimensioning of components is required for a reliable mimicking of the physiopathological behavior of the system. However, it is not possible to create a component for the afterload inertance, and inertance contributions are present in the entire circuit. Hence, in the literature, inertance is neglected or qualitatively evaluated. In this paper, we propose two quantitative methods (Maximum-likelihood estimation (MLE) and Bayesian estimation) for estimating afterload inertance based on observed pressure and flow waveforms. These methods are also applied to a real mock loop system. Results show that the system has an inertance comparable with the literature reference value of the entire systemic circulation, and that the expected variations over inlet average flow and pulse frequency are in general confirmed. Comparing the methods, the Bayesian approach results in higher and more stable estimations than the classical MLE.

Improving Blood Donor Care in a Collection Center Through Advanced Data Exploitation.

BACKGROUND: Blood collection centers can take advantage of the huge amount of data collected on donors over the years to predict and detect early the onset of several diseases, However, dedicated tools are needed to carry out these analyses. OBJECTIVES: This work develops a tool that combines available data with predictive tools to provide alerts to physicians and enable them to effectively visualize the history of critical donors in terms of the parameters that led to the alert. METHODS: The developed tool consists of data exchanging functions, interfaces to raise alerts and visualize donor history, and predictive algorithms. It was designed to be simple, modular and flexible. RESULTS: A prototype was applied to the Milan department of the Associazione Volontari Italiani Sangue, and was deemed suitable for prevention and early diagnosis objectives by the physicians of the center. The included Machine Learning predictive algorithms provided good estimates for the variables considered in the prototype. CONCLUSION: Prevention and early diagnosis activities in blood collection centers can be effectively supported by properly using and displaying donor clinical data through a dedicated software tool.

Surgical planning of arteriovenous fistulae in routine clinical practice: A machine learning predictive tool.

BACKGROUND: Arteriovenous fistula (AVF) is the preferred vascular access (VA) for hemodialysis, but it is associated with high non-maturation and failure rates. Predicting patient-specific AVF maturation and postoperative changes in blood flow volumes (BFVs) and vessel diameters is of fundamental importance to support the choice of optimal AVF location and improve VA survival. The goal of this study was to employ machine learning (ML) in order to give physicians a fast and easy-to-use tool that provides accurate patient-specific predictions, useful to make AVF surgical planning decisions. METHODS: We applied a set of ML approaches on a dataset of 156 patients. Both parametric and non-parametric ML approaches, taking preoperative data as input, were exploited to predict maturation, postoperative BFVs, and diameters. The best approach associated with lowest cross-validation errors between predictions and real measurements was then chosen to provide estimates and quantify prediction errors. RESULTS: The k-NN was the best approach to predict brachial BFV, AVF maturation, and other VA variables, and it was also associated with the least computational effort. With this approach, the confusion matrices proved the high accuracy of the prediction for AVF maturation (96.8%) and the low absolute error distribution for the continuous BFV and diameter variables. CONCLUSIONS: Our data-based approach provided accurate patient-specific predictions for different AVF configurations, requiring short computational time as compared to a physical model we previously developed. By supporting VA surgical planning, this fast computing approach could allow AVF surgical planning and help reducing the rate of non-maturation, which might ultimately have a broad impact on the management of hemodialysis patients.

Haemodynamic validation of the three-step HFA-PEFF algorithm to diagnose heart failure with preserved ejection fraction.

AIMS: The HFA-PEFF algorithm (Heart Failure Association-Pre-test assessment, Echocardiography and natriuretic peptide score, Functional testing in cases of uncertainty, Final aetiology) is a three-step algorithm to diagnose heart failure with preserved ejection fraction (HFpEF). It provides a three-level likelihood of HFpEF: low (score < 2), intermediate (score 2-4), or high (score > 4). HFpEF may be confirmed in individuals with a score > 4 (rule-in approach). The second step of the algorithm is based on echocardiographic features and natriuretic peptide levels. The third step implements diastolic stress echocardiography (DSE) for controversial diagnostic cases. We sought to validate the three-step HFA-PEFF algorithm against a haemodynamic diagnosis of HFpEF based on rest and exercise right heart catheterization (RHC). METHODS AND RESULTS: Seventy-three individuals with exertional dyspnoea underwent a full diagnostic work-up following the HFA-PEFF algorithm, including DSE and rest/exercise RHC. The association between the HFA-PEFF score and a haemodynamic diagnosis of HFpEF, as well as the diagnostic performance of the HFA-PEFF algorithm vs. RHC, was assessed. The diagnostic performance of left atrial (LA) strain < 24.5% and LA strain/E/E' < 3% was also assessed. The probability of HFpEF was low/intermediate/high in 8%/52%/40% of individuals at the second step of the HFA-PEFF algorithm and 8%/49%/43% at the third step. After RHC, 89% of patients were diagnosed as HFpEF and 11% as non-cardiac dyspnoea. The HFA-PEFF score resulted associated with the invasive haemodynamic diagnosis of HFpEF (P < 0.001). Sensitivity and specificity of the HFA-PEFF score for the invasive haemodynamic diagnosis of HFpEF were 45% and 100% for the second step of the algorithm and 46% and 88% for the third step of the algorithm. Neither age, sex, body mass index, obesity, chronic obstructive pulmonary disease, or paroxysmal atrial fibrillation influenced the performance of the HFA-PEFF algorithm, as these characteristics were similarly distributed over the true positive, true negative, false positive, and false negative cases. Sensitivity of the second step of the HFA-PEFF score was non-significantly improved to 60% (P = 0.08) by lowering the rule-in threshold to >3. LA strain alone had a sensitivity and specificity of 39% and 14% for haemodynamic HFpEF, increasing to 55% and 22% when corrected for E/E'. CONCLUSIONS: As compared with rest/exercise RHC, the HFA-PEFF score lacks sensitivity: Half of the patients were wrongly classified as non-cardiac dyspnoea after non-invasive tests, with a minimal impact of DSE in modifying HFpEF likelihood.

Optimal Robust Search for Parameter Values of Qualitative Models of Gene Regulatory Networks.

Computational and mathematical models are a must for the in silico analysis or design of Gene Regulatory Networks (GRN)as they offer a theoretical context to deeply address biological regulation. We have proposed a framework where models of network dynamics are expressed through a class of nonlinear and temporal multiscale Ordinary Differential Equations (ODE). To find out models that disclose network structures underlying an observed or desired network behavior, and parameter values that enable the candidate models to reproduce such behavior, we follow a reasoning cycle that alternates procedures for model selection and parameter refinement. Plausible network models are first selected via qualitative simulation, and next their parameters are given quantitative values such that the ODE model solution reproduces the specified behavior. This paper gives algorithms to tackle the parameter refinement problem formulated as an optimization problem. We search, within the parameter space symbolically expressed, for the largest hypersphere whose points correspond to parameter values such that the ODE solution gives an instance of the given qualitative trajectory. Our approach overcomes the limitation of a previously proposed stochastic approach, namely computational load and very reduced scalability. Its applicability and effectiveness are demonstrated through two benchmark synthetic networks with different complexity.

Automatic cyst and kidney segmentation in autosomal dominant polycystic kidney disease: Comparison of U-Net based methods.

Autosomal Dominant Polycystic Kidney Disease is a genetic disease that causes uncontrolled growth of fluid-filled cysts in the kidney. Kidney enlargement resulting from the expansion of cysts is continuous and often associated with decreased renal function and kidney failure. Mouse and rat models are necessary to discover new drugs able to halt the progression of the disease. The analysis of the effects of pharmacological interventions in these models is based on renal morphology and quantification of changes in total renal volume and cyst volume. This requires a proper, reproducible and fast segmentation of the kidney images. We propose a set of fully convolutional networks for kidney and cyst segmentation in micro-CT images, based on the U-Net architecture, to compare them and analyze which ones perform better on contrast-enhanced micro-CT images from normal rats and rats with Autosomal Dominant Polycystic Kidney Disease. Networks have been tested on a series images, and the performance has been evaluated in terms of Intersection over Union and Dice coefficients. Results showed that the best performing networks are the U-Net in which a batch normalization layer is applied after each pair of 3 × 3 convolutions, and the U-Net in which convolutional layers are replaced by inception blocks. Results also showed accurate cyst-to-kidney volume ratios obtained from the segmented images, which is one of main metrics of interest. Finally, segmentation performance has been found to be stable as the images in the training set vary. Therefore, the proposed automatic methodology is suitable and immediately applicable to segment cysts and kidney from micro-CT images, and directly provides the cyst-to-kidney volume ratio.

A patient-specific image-based approach to estimate pulmonary artery stiffness based on vessel constitutive model.

An accurate non-invasive evaluation of the mechanical properties of the vessel wall is important for a variety of screening protocols and surgical treatments. In this work, we focused on a section of the Pulmonary Artery (PA), and developed a patient-specific approach to estimate its stiffness in terms of the Young's modulus along the circumferential direction (E). First, we developed a patient-specific semi-automatic approach to estimate its expected value and standard deviation. To this end, pressure-length curves were derived from magnetic resonance images acquired during the cardiac cycle and information on vessel pressure obtained by catheterization. Then, the estimates of E were derived through a maximum likelihood estimation approach based on a vessel constitutive law. In particular, we analyzed the entire PA boundary and an arc free from surrounding organs. Second, we applied the approach to the study of pulmonary endarterectomy (PEA) for the treatment of chronic thromboembolic pulmonary hypertension (CTEPH). We observed a decrease in the circumferential E after PEA for the whole boundary, while no clear trend was observed for the free arc. The low standard deviations associated with the estimates showed high accuracy when considering the entire boundary, while greater variability was observed for the free arc, which was however limited. Finally, reliable hysteretic behavior was obtained from the reconstructed pressure-length curves.