Artificial intelligence to predict individualized outcome of acute ischemic stroke patients: The SIBILLA project.

INTRODUCTION: Formulating reliable prognosis for ischemic stroke patients remains a challenging task. We aimed to develop an artificial intelligence model able to formulate in the first 24 h after stroke an individualized prognosis in terms of NIHSS. PATIENTS AND METHODS: Seven hundred ninety four acute ischemic stroke patients were divided into a training (597) and testing (197) cohort. Clinical and instrumental data were collected in the first 24 h. We evaluated the performance of four machine-learning models (Random Forest, K-Nearest Neighbors, Support Vector Machine, XGBoost) in predicting NIHSS at discharge both in terms of variation between discharge and admission (regressor approach) and in terms of severity class namely NIHSS 0-5, 6-10, 11-20, >20 (classifier approach). We used Shapley Additive exPlanations values to weight features impact on predictions. RESULTS: XGBoost emerged as the best performing model. The classifier and regressor approaches perform similarly in terms of accuracy (80% vs 75%) and f1-score (79% vs 77%) respectively. However, the regressor has higher precision (85% vs 68%) in predicting prognosis of very severe stroke patients (NIHSS > 20). NIHSS at admission and 24 hours, GCS at 24 hours, heart rate, acute ischemic lesion on CT-scan and TICI score were the most impacting features on the prediction. DISCUSSION: Our approach, which employs an artificial intelligence based-tool, inherently able to continuously learn and improve its performance, could improve care pathway and support stroke physicians in the communication with patients and caregivers. CONCLUSION: XGBoost reliably predicts individualized outcome in terms of NIHSS at discharge in the first 24 hours after stroke.

Radiomics-enhanced early regression index for predicting treatment response in rectal cancer: a multi-institutional 0.35 T MRI-guided radiotherapy study.

PURPOSE: The accurate prediction of treatment response in locally advanced rectal cancer (LARC) patients undergoing MRI-guided radiotherapy (MRIgRT) is essential for optimising treatment strategies. This multi-institutional study aimed to investigate the potential of radiomics in enhancing the predictive power of a known radiobiological parameter (Early Regression Index, ERITCP) to evaluate treatment response in LARC patients treated with MRIgRT. METHODS: Patients from three international sites were included and divided into training and validation sets. 0.35 T T2*/T1-weighted MR images were acquired during simulation and at each treatment fraction. The biologically effective dose (BED) conversion was used to account for different radiotherapy schemes: gross tumour volume was delineated on the MR images corresponding to specific BED levels and radiomic features were then extracted. Multiple logistic regression models were calculated, combining ERITCP with other radiomic features. The predictive performance of the different models was evaluated on both training and validation sets by calculating the receiver operating characteristic (ROC) curves. RESULTS: A total of 91 patients was enrolled: 58 were used as training, 33 as validation. Overall, pCR was observed in 25 cases. The model showing the highest performance was obtained combining ERITCP at BED = 26 Gy with a radiomic feature (10th percentile of grey level histogram, 10GLH) calculated at BED = 40 Gy. The area under ROC curve (AUC) of this combined model was 0.98 for training set and 0.92 for validation set, significantly higher (p = 0.04) than the AUC value obtained using ERITCP alone (0.94 in training and 0.89 in validation set). CONCLUSION: The integration of the radiomic analysis with ERITCP improves the pCR prediction in LARC patients, offering more precise predictive models to further personalise 0.35 T MRIgRT treatments of LARC patients.

The Image Biomarker Standardization Initiative: Standardized Convolutional Filters for Reproducible Radiomics and Enhanced Clinical Insights.

Filters are commonly used to enhance specific structures and patterns in images, such as vessels or peritumoral regions, to enable clinical insights beyond the visible image using radiomics. However, their lack of standardization restricts reproducibility and clinical translation of radiomics decision support tools. In this special report, teams of researchers who developed radiomics software participated in a three-phase study (September 2020 to December 2022) to establish a standardized set of filters. The first two phases focused on finding reference filtered images and reference feature values for commonly used convolutional filters: mean, Laplacian of Gaussian, Laws and Gabor kernels, separable and nonseparable wavelets (including decomposed forms), and Riesz transformations. In the first phase, 15 teams used digital phantoms to establish 33 reference filtered images of 36 filter configurations. In phase 2, 11 teams used a chest CT image to derive reference values for 323 of 396 features computed from filtered images using 22 filter and image processing configurations. Reference filtered images and feature values for Riesz transformations were not established. Reproducibility of standardized convolutional filters was validated on a public data set of multimodal imaging (CT, fluorodeoxyglucose PET, and T1-weighted MRI) in 51 patients with soft-tissue sarcoma. At validation, reproducibility of 486 features computed from filtered images using nine configurations × three imaging modalities was assessed using the lower bounds of 95% CIs of intraclass correlation coefficients. Out of 486 features, 458 were found to be reproducible across nine teams with lower bounds of 95% CIs of intraclass correlation coefficients greater than 0.75. In conclusion, eight filter types were standardized with reference filtered images and reference feature values for verifying and calibrating radiomics software packages. A web-based tool is available for compliance checking.

A novel risk score predicting 30-day hospital re-admission of patients with acute stroke by machine learning model.

BACKGROUND: The 30-day hospital re-admission rate is a quality measure of hospital care to monitor the efficiency of the healthcare system. The hospital re-admission of acute stroke (AS) patients is often associated with higher mortality rates, greater levels of disability and increased healthcare costs. The aim of our study was to identify predictors of unplanned 30-day hospital re-admissions after discharge of AS patients and define an early re-admission risk score (RRS). METHODS: This observational, retrospective study was performed on AS patients who were discharged between 2014 and 2019. Early re-admission predictors were identified by machine learning models. The performances of these models were assessed by receiver operating characteristic curve analysis. RESULTS: Of 7599 patients with AS, 3699 patients met the inclusion criteria, and 304 patients (8.22%) were re-admitted within 30 days from discharge. After identifying the predictors of early re-admission by logistic regression analysis, RRS was obtained and consisted of seven variables: hemoglobin level, atrial fibrillation, brain hemorrhage, discharge home, chronic obstructive pulmonary disease, one and more than one hospitalization in the previous year. The cohort of patients was then stratified into three risk categories: low (RRS = 0-1), medium (RRS = 2-3) and high (RRS >3) with re-admission rates of 5%, 8% and 14%, respectively. CONCLUSIONS: The identification of risk factors for early re-admission after AS and the elaboration of a score to stratify at discharge time the risk of re-admission can provide a tool for clinicians to plan a personalized follow-up and contain healthcare costs.

Early treatment of type II SMA slows rate of progression of scoliosis.

BACKGROUND: Type II spinal muscular atrophy (SMA) often leads to scoliosis in up to 90% of cases. While pharmacological treatments have shown improvements in motor function, their impact on scoliosis progression remains unclear. This study aims to evaluate potential differences in scoliosis progression between treated and untreated SMA II patients. METHODS: Treatment effect on Cobb's angle annual changes and on reaching a 50° Cobb angle was analysed in treated and untreated type II SMA patients with a minimum 1.5-year follow-up. A sliding cut-off approach identified the optimal treatment subpopulation based on age, Cobb angle and Hammersmith Functional Motor Scale Expanded at the initial visit. Mann-Whitney U-test assessed statistical significance. RESULTS: There were no significant differences in baseline characteristics between the untreated (n=46) and treated (n=39) populations. The mean Cobb angle variation did not significantly differ between the two groups (p=0.4). Optimal cut-off values for a better outcome were found to be having a Cobb angle <26° or an age <4.5 years. When using optimal cut-off, the treated group showed a lower mean Cobb variation compared with the untreated group (5.61 (SD 4.72) degrees/year vs 10.05 (SD 6.38) degrees/year; p=0.01). Cox-regression analysis indicated a protective treatment effect in reaching a 50° Cobb angle, significant in patients <4.5 years old (p=0.016). CONCLUSION: This study highlights that pharmacological treatment, if initiated early, may slow down the progression of scoliosis in type II SMA patients. Larger studies are warranted to further investigate the effectiveness of individual pharmacological treatment on scoliosis progression in this patient population.

The future of neonatal lung ultrasound: Validation of an artificial intelligence model for interpreting lung scans. A multicentre prospective diagnostic study.

BACKGROUND: Artificial intelligence (AI) is a promising field in the neonatal field. We focused on lung ultrasound (LU), a useful tool for the neonatologist. Our aim was to train a neural network to create a model able to interpret LU. METHODS: Our multicentric, prospective study included newborns with gestational age (GA) ≥ 33 + 0 weeks with early tachypnea/dyspnea/oxygen requirements. For each baby, three LU were performed: within 3 h of life (T0), at 4-6 h of life (T1), and in the absence of respiratory support (T2). Each scan was processed to extract the region of interest used to train a neural network to classify it according to the LU score (LUS). We assessed sensitivity, specificity, positive and negative predictive value of the AI model's scores in predicting the need for respiratory assistance with nasal continuous positive airway pressure and for surfactant, compared to an already studied and established LUS. RESULTS: We enrolled 62 newborns (GA = 36 ± 2 weeks). In the prediction of the need for CPAP, we found a cutoff of 6 (at T0) and 5 (at T1) for both the neonatal lung ultrasound score (nLUS) and AI score (AUROC 0.88 for T0 AI model, 0.80 for T1 AI model). For the outcome "need for surfactant therapy", results in terms of area under receiver operator characteristic (AUROC) are 0.84 for T0 AI model and 0.89 for T1 AI model. In the prediction of surfactant therapy, we found a cutoff of 9 for both scores at T0, at T1 the nLUS cutoff was 6, while the AI's one was 5. Classification accuracy was good both at the image and class levels. CONCLUSIONS: This is, to our knowledge, the first attempt to use an AI model to interpret early neonatal LUS and can be extremely useful for neonatologists in the clinical setting.

A process mining approach for clinical guidelines compliance: real-world application in rectal cancer.

In the era of evidence-based medicine, several clinical guidelines were developed, supporting cancer management from diagnosis to treatment and aiming to optimize patient care and hospital resources. Nevertheless, individual patient characteristics and organizational factors may lead to deviations from these standard recommendations during clinical practice. In this context, process mining in healthcare constitutes a valid tool to evaluate conformance of real treatment pathways, extracted from hospital data warehouses as event log, to standard clinical guidelines, translated into computer-interpretable formats. In this study we translate the European Society of Medical Oncology guidelines for rectal cancer treatment into a computer-interpretable format using Pseudo-Workflow formalism (PWF), a language already employed in pMineR software library for Process Mining in Healthcare. We investigate the adherence of a real-world cohort of rectal cancer patients treated at Fondazione Policlinico Universitario A. Gemelli IRCCS, data associated with cancer diagnosis and treatment are extracted from hospital databases in 453 patients diagnosed with rectal cancer. PWF enables the easy implementation of guidelines in a computer-interpretable format and visualizations that can improve understandability and interpretability of physicians. Results of the conformance checking analysis on our cohort identify a subgroup of patients receiving a long course treatment that deviates from guidelines due to a moderate increase in radiotherapy dose and an addition of oxaliplatin during chemotherapy treatment. This study demonstrates the importance of PWF to evaluate clinical guidelines adherence and to identify reasons of deviations during a treatment process in a real-world and multidisciplinary setting.

Eligibility for the 4 Pharmacological Pillars in Heart Failure With Reduced Ejection Fraction at Discharge.

Background Guidelines recommend using multiple drugs in patients with heart failure (HF) with reduced ejection fraction, but there is a paucity of real-world data on the simultaneous initiation of the 4 pharmacological pillars at discharge after a decompensation event. Methods and Results A retrospective data mart, including patients diagnosed with HF, was implemented. Consecutively admitted patients with HF with reduced ejection fraction were selected through an automated approach and categorized according to the number/type of treatments prescribed at discharge. The prevalence of contraindications and cautions for HF with reduced ejection fraction treatments was systematically assessed. Logistic regression models were fitted to assess predictors of the number of treatments (≥2 versus <2 drugs) prescribed and the risk of rehospitalization. A population of 305 patients with a first episode of HF hospitalization and a diagnosis of HF with reduced ejection fraction (ejection fraction, <40%) was selected. At discharge, 49.2% received 2 current recommended drugs, ß-blockers were prescribed in 93.4%, while a renin-angiotensin system inhibitor or an angiotensin receptor-neprilysin inhibitor was prescribed in 68.2%. A mineralocorticoid receptor antagonist was prescribed in 32.5%, although none of the patients showed contraindications to mineralocorticoid receptor antagonist prescription. A sodium-glucose cotransporter 2 inhibitor could be prescribed in 71.1% of patients. On the basis of current recommendations, 46.2% could receive the 4 foundational drugs at discharge. Renal dysfunction was associated with <2 foundational drugs prescribed. After adjusting for age and renal function, use of ≥2 drugs was associated with lower risk of rehospitalization during the 30 days after discharge. Conclusions A quadruple therapy could be directly implementable at discharge, potentially providing prognostic advantages. Renal dysfunction was the main prevalent condition limiting this approach.

GENERATOR HEART FAILURE DataMart: An integrated framework for heart failure research.

Background: Heart failure (HF) is a multifaceted clinical syndrome characterized by different etiologies, risk factors, comorbidities, and a heterogeneous clinical course. The current model, based on data from clinical trials, is limited by the biases related to a highly-selected sample in a protected environment, constraining the applicability of evidence in the real-world scenario. If properly leveraged, the enormous amount of data from real-world may have a groundbreaking impact on clinical care pathways. We present, here, the development of an HF DataMart framework for the management of clinical and research processes. Methods: Within our institution, Fondazione Policlinico Universitario A. Gemelli in Rome (Italy), a digital platform dedicated to HF patients has been envisioned (GENERATOR HF DataMart), based on two building blocks: 1. All retrospective information has been integrated into a multimodal, longitudinal data repository, providing in one single place the description of individual patients with drill-down functionalities in multiple dimensions. This functionality might allow investigators to dynamically filter subsets of patient populations characterized by demographic characteristics, biomarkers, comorbidities, and clinical events (e.g., re-hospitalization), enabling agile analyses of the outcomes by subsets of patients. 2. With respect to expected long-term health status and response to treatments, the use of the disease trajectory toolset and predictive models for the evolution of HF has been implemented. The methodological scaffolding has been constructed in respect of a set of the preferred standards recommended by the CODE-EHR framework. Results: Several examples of GENERATOR HF DataMart utilization are presented as follows: to select a specific retrospective cohort of HF patients within a particular period, along with their clinical and laboratory data, to explore multiple associations between clinical and laboratory data, as well as to identify a potential cohort for enrollment in future studies; to create a multi-parametric predictive models of early re-hospitalization after discharge; to cluster patients according to their ejection fraction (EF) variation, investigating its potential impact on hospital admissions. Conclusion: The GENERATOR HF DataMart has been developed to exploit a large amount of data from patients with HF from our institution and generate evidence from real-world data. The two components of the HF platform might provide the infrastructural basis for a combined patient support program dedicated to continuous monitoring and remote care, assisting patients, caregivers, and healthcare professionals.

Preoperative Tumor Texture Analysis on MRI for High-Risk Disease Prediction in Endometrial Cancer: A Hypothesis-Generating Study.

OBJECTIVE: To develop and validate magnetic resonance (MR) imaging-based radiomics models for high-risk endometrial cancer (EC) prediction preoperatively, to be able to estimate deep myometrial invasion (DMI) and lymphovascular space invasion (LVSI), and to discriminate between low-risk and other categories of risk as proposed by ESGO/ESTRO/ESP (European Society of Gynaecological Oncology-European Society for Radiotherapy & Oncology and European Society of Pathology) guidelines. METHODS: This retrospective study included 96 women with EC who underwent 1.5-T MR imaging before surgical staging between April 2009 and May 2019 in two referral centers divided into training (T = 73) and validation cohorts (V = 23). Radiomics features were extracted using the MODDICOM library with manual delineation of whole-tumor volume on MR images (axial T2-weighted). Diagnostic performances of radiomic models were evaluated by area under the receiver operating characteristic (ROC) curve in training (AUCT) and validation (AUCV) cohorts by using a subset of the most relevant texture features tested individually in univariate analysis using Wilcoxon-Mann-Whitney. RESULTS: A total of 228 radiomics features were extracted and ultimately limited to 38 for DMI, 29 for LVSI, and 15 for risk-classes prediction for logistic radiomic modeling. Whole-tumor radiomic models yielded an AUCT/AUCV of 0.85/0.68 in DMI estimation, 0.92/0.81 in LVSI prediction, and 0.84/0.76 for differentiating low-risk vs other risk classes (intermediate/high-intermediate/high). CONCLUSION: MRI-based radiomics has great potential in developing advanced prognostication in EC.

Deep-Learning to Predict BRCA Mutation and Survival from Digital H&E Slides of Epithelial Ovarian Cancer.

BRCA 1/2 genes mutation status can already determine the therapeutic algorithm of high grade serous ovarian cancer patients. Nevertheless, its assessment is not sufficient to identify all patients with genomic instability, since BRCA 1/2 mutations are only the most well-known mechanisms of homologous recombination deficiency (HR-d) pathway, and patients displaying HR-d behave similarly to BRCA mutated patients. HRd assessment can be challenging and is progressively overcoming BRCA testing not only for prognostic information but more importantly for drugs prescriptions. However, HR testing is not already integrated in clinical practice, it is quite expensive and it is not refundable in many countries. Selecting patients who are more likely to benefit from this assessment (BRCA 1/2 WT patients) at an early stage of the diagnostic process, would allow an optimization of genomic profiling resources. In this study, we sought to explore whether somatic BRCA1/2 genes status can be predicted using computational pathology from standard hematoxylin and eosin histology. In detail, we adopted a publicly available, deep-learning-based weakly supervised method that uses attention-based learning to automatically identify sub regions of high diagnostic value to accurately classify the whole slide (CLAM). The same model was also tested for progression free survival (PFS) prediction. The model was tested on a cohort of 664 (training set: n = 464, testing set: n = 132) ovarian cancer patients, of whom 233 (35.1%) had a somatic BRCA 1/2 mutation. An area under the curve of 0.7 and 0.55 was achieved in the training and testing set respectively. The model was then further refined by manually identifying areas of interest in half of the cases. 198 images were used for training (126/72) and 87 images for validation (55/32). The model reached a zero classification error on the training set, but the performance was 0.59 in terms of validation ROC AUC, with a 0.57 validation accuracy. Finally, when applied to predict PFS, the model achieved an AUC of 0.71, with a negative predictive value of 0.69, and a positive predictive value of 0.75. Based on these analyses, we have planned further steps of development such as proving a reference classification performance, exploring the hyperparameters space for training optimization, eventually tweaking the learning algorithms and the neural networks architecture for better suiting this specific task. These actions may allow the model to improve performances for all the considered outcomes.

A deep learning approach to generate synthetic CT in low field MR-guided radiotherapy for lung cases.

INTRODUCTION: This study aims to apply a conditional Generative Adversarial Network (cGAN) to generate synthetic Computed Tomography (sCT) from 0.35 Tesla Magnetic Resonance (MR) images of the thorax. METHODS: Sixty patients treated for lung lesions were enrolled and divided into training (32), validation (8), internal (10,TA) and external (10,TB) test set. Image accuracy of generated sCT was evaluated computing the mean absolute (MAE) and mean error (ME) with respect the original CT. Three treatment plans were calculated for each patient considering MRI as reference image: original CT, sCT (pure sCT) and sCT with GTV density override (hybrid sCT) were used as Electron Density (ED) map. Dose accuracy was evaluated comparing treatment plans in terms of gamma analysis and Dose Volume Histogram (DVH) parameters. RESULTS: No significant difference was observed between the test sets for image and dose accuracy parameters. Considering the whole test cohort, a MAE of 54.9 ± 10.5 HU and a ME of 4.4 ± 7.4 HU was obtained. Mean gamma passing rates for 2%/2mm, and 3%/3mm tolerance criteria were 95.5 ± 5.9% and 98.2 ± 4.1% for pure sCT, 96.1 ± 5.1% and 98.5 ± 3.9% for hybrid sCT: the difference between the two approaches was significant (p = 0.01). As regards DVH analysis, differences in target parameters estimation were found to be within 5% using hybrid approach and 20% using pure sCT. CONCLUSION: The DL algorithm here presented can generate sCT images in the thorax with good image and dose accuracy, especially when the hybrid approach is used. The algorithm does not suffer from inter-scanner variability, making feasible the implementation of MR-only workflows for palliative treatments.

Correction: Rizzo et al. Artificial Intelligence and OCT Angiography in Full Thickness Macular Hole. New Developments for Personalized Medicine. Diagnostics 2021, 11, 2319.

In the original publication [...].

Age, corticosteroid treatment and site of mutations affect motor functional changes in young boys with Duchenne Muscular Dystrophy.

The aim of this study was to establish the possible effect of age, corticosteroid treatment and brain dystrophin involvement on motor function in young boys affected by Duchenne Muscular Dystrophy who were assessed using the North Star Ambulatory Assessment between the age of 4 and 7 years. The study includes 951 North Star assessments from 226 patients. Patients were subdivided according to age, to the site of mutation and therefore to the involvement of different brain dystrophin isoforms and to corticosteroids duration. There was a difference in the maximum North Star score achieved among patients with different brain dystrophin isoforms (p = 0.007). Patients with the involvement of Dp427, Dp140 and Dp71, had lower maximum NSAA scores when compared to those with involvement of Dp427 and Dp140 or of Dp427 only. The difference in the age when the maximum score was achieved in the different subgroups did not reach statistical significance. Using a linear regression model on all assessments we found that each of the three variables, age, site of mutation and corticosteroid treatment had an influence on the NSAA values and their progression over time. A second analysis, looking at 12-month changes showed that within this time interval the magnitude of changes was related to corticosteroid treatment but not to site of mutation. Our findings suggest that each of the considered variables appear to play a role in the progression of North Star scores in patients between the age of 4 and 7 years and that these should be carefully considered in the trial design of boys in this age range.

Early lung ultrasound score to predict noninvasive ventilation needing in neonates from 33 weeks of gestational age: A multicentric study.

OBJECTIVE: To propose an early lung ultrasound (LUS) score for the prediction of the need for respiratory assistance in newborns of gestational age (GA) ≥ 33 weeks presenting respiratory distress. STUDY DESIGN AND SETTING: Multicenter, prospective observational study in third-level neonatal intensive care units. PATIENT SELECTION: Infants with GA ≥ 33 + 0 weeks with respiratory distress within 3 h of life. METHODS: Three LUS for each patient were collected: within 3 h of life (T0), at 4-6 h of life (T1), and at the resolution of symptoms (T2). The primary aim was to assess the validity of the early LUS score in predicting the need for continuous positive airway pressure (CPAP). We also evaluated the validity of the score in predicting the need for surfactant, the scores' trend in our population, and any correlation with the duration of ventilation and oxygen therapy. RESULTS: Sixty-two patients were enrolled in the study. The mean GA was 36 weeks. The receiver operating characteristic analysis for the LUS T0 and T1 yielded area under the curves of 0.91 and 0.82 in predicting the need for CPAP, respectively. LUS score cut off of 6 (sensitivity 84.8%, specificity 86.2%) and 5 (sensitivity 66.7%, specificity 100%) were calculated at T0 and T1, respectively. We found significant correlations between LUS score and respiratory assistance, surfactant administration, and SpO2 /FiO2 ratio. CONCLUSION: An early LUS score is a good noninvasive predictor of the need for respiratory assistance with CPAP and surfactant administration in newborns with GA ≥ 33 weeks.

CT-Based Radiomics and Deep Learning for BRCA Mutation and Progression-Free Survival Prediction in Ovarian Cancer Using a Multicentric Dataset.

PURPOSE: Build predictive radiomic models for early relapse and BRCA mutation based on a multicentric database of high-grade serous ovarian cancer (HGSOC) and validate them in a test set coming from different institutions. METHODS: Preoperative CTs of patients with HGSOC treated at four referral centers were retrospectively acquired and manually segmented. Hand-crafted features and deep radiomics features were extracted respectively by dedicated software (MODDICOM) and a dedicated convolutional neural network (CNN). Features were selected with and without prior harmonization (ComBat harmonization), and models were built using different machine learning algorithms, including clinical variables. RESULTS: We included 218 patients. Radiomic models showed low performance in predicting both BRCA mutation (AUC in test set between 0.46 and 0.59) and 1-year relapse (AUC in test set between 0.46 and 0.56); deep learning models demonstrated similar results (AUC in the test of 0.48 for BRCA and 0.50 for relapse). The inclusion of clinical variables improved the performance of the radiomic models to predict BRCA mutation (AUC in the test set of 0.74). CONCLUSIONS: In our multicentric dataset, representative of a real-life clinical scenario, we could not find a good radiomic predicting model for PFS and BRCA mutational status, with both traditional radiomics and deep learning, but the combination of clinical and radiomic models improved model performance for the prediction of BRCA mutation. These findings highlight the need for standardization through the whole radiomic pipelines and robust multicentric external validations of results.

CT angiography-based radiomics as a tool for carotid plaque characterization: a pilot study.

PURPOSES: Radiomics is a quantitative method able to analyze a high-throughput extraction of minable imaging features. Herein, we aim to develop a CT angiography-based radiomics analysis and machine learning model for carotid plaques to discriminate vulnerable from no vulnerable plaques. MATERIALS AND METHODS: Thirty consecutive patients with carotid atherosclerosis were enrolled in this pilot study. At surgery, a binary classification of plaques was adopted ("hard" vs "soft"). Feature extraction was performed using the R software package Moddicom. Pairwise feature interdependencies were evaluated using the Spearman rank correlation coefficient. A univariate analysis was performed to assess the association between each feature and the plaque classification and chose top-ranked features. The feature predictive value was investigated using binary logistic regression. A stepwise backward elimination procedure was performed to minimize the Akaike information criterion (AIC). The final significant features were used to build the models for binary classification of carotid plaques, including logistic regression (LR), support vector machine (SVM), and classification and regression tree analysis (CART). All models were cross-validated using fivefold cross validation. Class-specific accuracy, precision, recall and F-measure evaluation metrics were used to quantify classifier output quality. RESULTS: A total of 230 radiomics features were extracted from each plaque. Pairwise Spearman correlation between features reported a high level of correlations, with more than 80% correlating with at least one other feature at |ρ|> 0.8. After a stepwise backward elimination procedure, the entropy and volume features were found to be the most significantly associated with the two plaque groups (p < 0.001), with AUCs of 0.92 and 0.96, respectively. The best performance was registered by the SVM classifier with the RBF kernel, with accuracy, precision, recall and F-score equal to 86.7, 92.9, 81.3 and 86.7%, respectively. The CART classification tree model for the entropy and volume features model achieved 86.7% well-classified plaques and an AUC of 0.987. CONCLUSION: This pilot study highlighted the potential of CTA-based radiomics and machine learning to discriminate plaque composition. This new approach has the potential to provide a reliable method to improve risk stratification in patients with carotid atherosclerosis.

Predictive models in SMA II natural history trajectories using machine learning: A proof of concept study.

It is known from previous literature that type II Spinal Muscular Atrophy (SMA) patients generally, after the age of 5 years, presents a steep deterioration until puberty followed by a relative stability, as most abilities have been lost. Although it is possible to identify points of slope indicating early improvement, steep decline and relative stabilizations, there is still a lot of variability within each age group and it's not always possible to predict individual trajectories of progression from age only. The aim of the study was to develop a predictive model based on machine learning using an XGBoost algorithm for regression and report, explore and quantify, in a single centre longitudinal natural history study, the influence of clinical variables on the 6/12-months Hammersmith Motor Functional Scale Expanded score prediction (HFMSE). This study represents the first approach to artificial intelligence and trained models for the prediction of individualized trajectories of HFMSE disease progression using individual characteristics of the patient. The application of this method to larger cohorts may allow to identify different classes of progression, a crucial information at the time of the new commercially available therapies.

CT-based radiomics modeling for skull dysmorphology severity and surgical outcome prediction in children with isolated sagittal synostosis: a hypothesis-generating study.

PURPOSE: To investigate the potentialities of radiomic analysis and develop radiomic models to predict the skull dysmorphology severity and post-surgical outcome in children with isolated sagittal synostosis (ISS). MATERIALS AND METHODS: Preoperative high-resolution CT scans of infants with ISS treated with surgical correction were retrospectively reviewed. The sagittal suture (ROI_entire) and its sections (ROI_anterior/central/posterior) were segmented. Radiomic features extracted from ROI_entire were correlated to the scaphocephalic severity, while radiomic features extracted from ROI_anterior/central/posterior were correlated to the post-surgical outcome. Logistic regression models were built from selected radiomic features and validated to predict the scaphocephalic severity and post-surgical outcome. RESULTS: A total of 105 patients were enrolled in this study. The kurtosis was obtained from the feature selection process for both scaphocephalic severity and post-surgical outcome prediction. The model predicting the scaphocephalic severity had an area under the curve (AUC) of the receiver operating characteristic of 0.71 and a positive predictive value of 0.83 for the testing set. The model built for the post-surgical outcome showed an AUC (95% CI) of 0.75 (0.61;0.88) and a negative predictive value (95% CI) of 0.95 (0.84;0.99). CONCLUSION: Our results suggest that radiomics could be useful in quantifying tissue microarchitecture along the mid-suture space and potentially provide relevant biological information about the sutural ossification processes to predict the onset of skull deformities and stratify post-surgical outcome.

Development of a prognostic model of overall survival in oropharyngeal cancer from real-world data: PRO.M.E.THE.O.

Objective: The PRO.M.E.THE.O. study (PredictiOn Models in Ent cancer for anti-EGFR based THErapy Optimization) aimed to develop a predictive model (PM) of overall survival (OS) for patients with locally advanced oropharyngeal cancer (LAOC) treated with radiotherapy (RT) and cetuximab (Cet) from an Italian dataset. Methods: We enrolled patients with LAOC from 6 centres treated with RT-Cet. Clinical and treatment variables were collected. Patients were randomly divided into training (TS) (80%) and validation (VS) (20%) sets. A binary logistic regression model was used on the TS with stepwise feature selection and then on VS. Timepoints of 2, 3 and 5 years were considered. The area under the curve (AUC) of receiver operating characteristic of 2, 3 and 5 year and confusion matrix statistics at 5-threshold were used as performance criteria. Results: Overall, 218 patients were enrolled and 174 (79.8%) were analysed. Age at diagnosis, gender, ECOG performance, clinical stage, dose to high-risk volume, overall treatment time and day of RT interruption were considered in the final PMs. The PMs were developed and represented by nomograms with AUC of 0.75, 0.73 and 0.73 for TS and 0.713, 0.713, 0.775 for VS at 2, 3 and 5 years, respectively. Conclusions: PRO.M.E.THE.O. allows the creation of a PM for OS in patients with LAOC treated with RT-Cet.