RESUMEN
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.
Asunto(s)
Procesamiento de Imagen Asistido por Computador , Radiómica , Humanos , Reproducibilidad de los Resultados , Biomarcadores , Imagen MultimodalRESUMEN
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.
Asunto(s)
Radiómica , Neoplasias del Recto , Humanos , Neoplasias del Recto/diagnóstico por imagen , Neoplasias del Recto/radioterapia , Neoplasias del Recto/patología , Imagen por Resonancia Magnética/métodos , Recto , Terapia Neoadyuvante/métodos , Estudios RetrospectivosRESUMEN
The role of the radiation oncologist in the management of patients affected by prostate cancer is increasingly considered thanks to important technological innovations that have marked the radiotherapeutic approach in its three main fields: external beam radiotherapy (EB-RT), brachytherapy (interventional radiotherapy, I-RT), and metabolic radiotherapy (M-RT) through the use of new radiopharmaceuticals. Regarding the modern brachytherapy, the introduction of intensity-modulated techniques (IM-IRT), thanks to the implementation of HDR remote-after loading machines, and image-guided techniques (IG-IRT), has led to advantages in optimizing dose distribution after implantation with the possibility of modulating the dose according to the intraprostatic dominant lesions, limiting the dose to the surrounding tissues with improvement in local control and a significant reduction in side effects. I-RT today represents a safe, scientifically established, effective and well-tolerated treatment for patients affected by prostate cancer. Like most special techniques, in order to obtain the best results, it must be performed in centers with a high volume of activity and consolidated experience with an interdisciplinary approach.
Asunto(s)
Braquiterapia , Neoplasias de la Próstata , Masculino , Humanos , Oncólogos de Radiación , Braquiterapia/efectos adversos , Braquiterapia/métodos , Neoplasias de la Próstata/patología , Dosificación RadioterapéuticaRESUMEN
BACKGROUND: Post-traumatic growth (PTG) is defined as "positive psychological change experienced as a result of the struggle with highly challenging life circumstances". Diagnosis of cancer leads to many psychological challenges. The recent pandemic forced oncological patients to face other multiple stressors. Resilience is a target of interest for PTG. The aim of this study is to analyze relationships between cancer trauma, COVID-19 pandemic stress, PTG and resilience over time. PARTICIPANTS AND PROCEDURE: One hundred forty-six patients (124 females, 22 males) in active oncological treatment were enrolled from September 2020: 45.2% (n = 66) diagnosed with gynecological cancer, 23.3% (n = 34) with breast cancer, 15.1% (n = 22) with lung cancer, 16.5% (n = 24) with other cancers. We conducted a prospective longitudinal study on oncological patients evaluated at: diagnosis (T0), 6 (T1) and 12 months (T2) by means of the following self-administered tests: Distress Thermometer (DT), Hospital Anxiety and Depression Scale (HADS), Impact of Event Scale Revised (IES-R), Post-traumatic Growth Inventory (PTGI), Perceived Stress Scale (PSS), Connor-Davidson Resilience Scale (CD-RISC). RESULTS: DT decreased over time (T0 vs. T2, p < .001). HADS decreased from T0 to T2 (p < .001). The PTG subscales regarding new possibilities and appreciating life improved comparing T0 vs. T2 (p = .029; p = .013), as well as the total index of PTG (p = .027). The IES avoidance subscale score decreased over time (T0 vs. T1, p = .035). CONCLUSIONS: For some patients, the cancer experience is characterized not only by psychological distress but also by the presence and growth of positive aspects, such as the tendency to positively reconsider the value and importance of life, health and social relationships.
RESUMEN
The aim of the study was to build a machine learning-based predictive model to discriminate between hospitalized patients at low risk and high risk of bloodstream infection (BSI). A Data Mart including all patients hospitalized between January 2016 and December 2019 with suspected BSI was built. Multivariate logistic regression was applied to develop a clinically interpretable machine learning predictive model. The model was trained on 2016-2018 data and tested on 2019 data. A feature selection based on a univariate logistic regression first selected candidate predictors of BSI. A multivariate logistic regression with stepwise feature selection in five-fold cross-validation was applied to express the risk of BSI. A total of 5660 hospitalizations (4026 and 1634 in the training and the validation subsets, respectively) were included. Eleven predictors of BSI were identified. The performance of the model in terms of AUROC was 0.74. Based on the interquartile predicted risk score, 508 (31.1%) patients were defined as being at low risk, 776 (47.5%) at medium risk, and 350 (21.4%) at high risk of BSI. Of them, 14.2% (72/508), 30.8% (239/776), and 64% (224/350) had a BSI, respectively. The performance of the predictive model of BSI is promising. Computational infrastructure and machine learning models can help clinicians identify people at low risk for BSI, ultimately supporting an antibiotic stewardship approach.
RESUMEN
BACKGROUND: Accurate nodal restaging is becoming clinically more important in patients with locally advanced rectal cancer (LARC) with the emergence of organ-preserving treatment after a good response to neoadjuvant chemoradiotherapy (nCRT). PURPOSE: To evaluate the accuracy of MRI in identifying negative N status (ypN0 patients) in LARC after nCRT. MATERIAL AND METHODS: 191 patients with LARC underwent MRI before and 6-8â¯weeks after nCRT and subsequent total mesorectal excision. Short-axis diameter of mesorectal lymph nodes was evaluated on the high resolution T2-weighted images to compare MRI restaging with histopathology.. RESULTS: 146 and 45 patients had a negative N status (ypN0) and positive N status (ypNâ¯+â¯), respectively. On restaging MRI, the 70â¯% reduction in size of the largest node was associated with an area under the curve (AUC) of 0.818 to predict ypN0 stage, with a sensitivity of 93.3â¯% and a negative predictive value (NPV) of 95.4â¯%. No nodes were observed in 38 pts (37 pts ypN0 and 1 patient ypNâ¯+â¯), with sensitivity and NPV of nodes disappearance for ypN0 stage of 93.3â¯% and 92.5â¯% respectively. A 2.2â¯mm cut-off in short-axis diameter was associated with an AUC of 0.83 for the prediction of ypN0 nodal stage, with sensitivity and NPV of 79,5% and 91.1â¯% respectively. CONCLUSION: A reduction in size of 70â¯% of the largest limph-node on MRI at rectal cancer restaging has high sensitivity and NPV for prediction of ypN0 stage after nCRT. The high NPV of node disappearance and of aâ¯≤â¯2.2â¯mm short-axis diameter is confirmed.
Asunto(s)
Imagen por Resonancia Magnética , Neoplasias del Recto , Humanos , Estadificación de Neoplasias , Imagen por Resonancia Magnética/métodos , Quimioradioterapia/métodos , Neoplasias del Recto/terapia , Neoplasias del Recto/tratamiento farmacológico , Terapia Neoadyuvante/métodos , Ganglios Linfáticos/diagnóstico por imagen , Ganglios Linfáticos/patología , Estudios RetrospectivosRESUMEN
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.
RESUMEN
BACKGROUND AND PURPOSE: Artificial Intelligence (AI) models in radiation therapy are being developed with increasing pace. Despite this, the radiation therapy community has not widely adopted these models in clinical practice. A cohesive guideline on how to develop, report and clinically validate AI algorithms might help bridge this gap. METHODS AND MATERIALS: A Delphi process with all co-authors was followed to determine which topics should be addressed in this comprehensive guideline. Separate sections of the guideline, including Statements, were written by subgroups of the authors and discussed with the whole group at several meetings. Statements were formulated and scored as highly recommended or recommended. RESULTS: The following topics were found most relevant: Decision making, image analysis, volume segmentation, treatment planning, patient specific quality assurance of treatment delivery, adaptive treatment, outcome prediction, training, validation and testing of AI model parameters, model availability for others to verify, model quality assurance/updates and upgrades, ethics. Key references were given together with an outlook on current hurdles and possibilities to overcome these. 19 Statements were formulated. CONCLUSION: A cohesive guideline has been written which addresses main topics regarding AI in radiation therapy. It will help to guide development, as well as transparent and consistent reporting and validation of new AI tools and facilitate adoption.
Asunto(s)
Inteligencia Artificial , Técnica Delphi , Humanos , Planificación de la Radioterapia Asistida por Computador/normas , Planificación de la Radioterapia Asistida por Computador/métodos , Oncología por Radiación/normas , Radioterapia/normas , Radioterapia/métodos , AlgoritmosRESUMEN
BACKGROUND: Myocardial injury is prevalent among patients hospitalized for COVID-19. However, the role of COVID-19 vaccines in modifying the risk of myocardial injury is unknown. OBJECTIVES: To assess the role of vaccines in modifying the risk of myocardial injury in COVID-19. METHODS: We enrolled COVID-19 patients admitted from March 2021 to February 2022 with known vaccination status and ≥1 assessment of hs-cTnI within 30 days from the admission. The primary endpoint was the occurrence of myocardial injury (hs-cTnI levels >99th percentile upper reference limit). RESULTS: 1019 patients were included (mean age 67.7±14.8 years, 60.8% male, 34.5% vaccinated against COVID-19). Myocardial injury occurred in 145 (14.2%) patients. At multivariate logistic regression analysis, advanced age, chronic kidney disease and hypertension, but not vaccination status, were independent predictors of myocardial injury. In the analysis according to age tertiles distribution, myocardial injury occurred more frequently in the III tertile (≥76 years) compared to other tertiles (I tertile:≤60 years;II tertile:61-75 years) (p<0.001). Moreover, in the III tertile, vaccination was protective against myocardial injury (OR 0.57, CI 95% 0.34-0.94; p=0.03), while a previous history of coronary artery disease was an independent positive predictor. In contrast, in the I tertile, chronic kidney disease (OR 6.94, 95% CI 1.31-36.79, p=0.02) and vaccination (OR 4.44, 95% CI 1.28-15.34, p=0.02) were independent positive predictors of myocardial injury. CONCLUSIONS: In patients ≥76 years, COVID-19 vaccines were protective for the occurrence of myocardial injury, while in patients ≤60 years, myocardial injury was associated with previous COVID-19 vaccination. Further studies are warranted to clarify the underlying mechanisms.