Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 49 - 58
Опубликована: Янв. 1, 2024
Язык: Английский
Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 49 - 58
Опубликована: Янв. 1, 2024
Язык: Английский
Life, Год журнала: 2024, Номер 14(6), С. 726 - 726
Опубликована: Июнь 3, 2024
Purpose: To evaluate the role of radiomics in preoperative outcome prediction cirrhotic patients who underwent transjugular intrahepatic portosystemic shunt (TIPS) using “controlled expansion covered stents”. Materials and Methods: This retrospective institutional review board-approved study included undergoing TIPS with controlled stent placement. From CT images, whole liver was segmented into Volumes Interest (VOIs) at unenhanced portal venous phase. Radiomics features were extracted, collected, analyzed. Subsequently, receiver operating characteristic (ROC) curves drawn to assess which could predict patients’ outcomes. The endpoints studied 6-month overall survival (OS), development hepatic encephalopathy (HE), grade II or higher HE according West Haven Criteria, clinical response, defined as absence rebleeding ascites. A radiomic model for then designed. Results: total 76 consecutive creation enrolled. highest performances terms area under curve (AUROC) observed “clinical response” “survival 6 months” 0.755 0.767, phase, respectively. Specifically, on basal scans, accuracy, specificity, sensitivity 66.42%, 63.93%, 73.75%, At an accuracy 65.34%, a specificity 62.38%, 74.00% demonstrated. Conclusions: pre-interventional machine learning-based algorithm be useful predicting response after patients.
Язык: Английский
Процитировано
1Journal of Medical Signals & Sensors, Год журнала: 2024, Номер 14(10)
Опубликована: Окт. 1, 2024
In this study, we want to evaluate the response Lutetium-177 (
Язык: Английский
Процитировано
1Deleted Journal, Год журнала: 2024, Номер 8(1)
Опубликована: Ноя. 29, 2024
Abstract Background Indices of tumor heterogeneity on somatostatin receptor PET/CT scans may potentially serve as predictive biomarkers treatment efficacy in neuroendocrine (NET) patients undergoing [ 177 Lu]Lu-DOTA-TATE PRRT. Methods NET who underwent therapy at the University Iowa from August 2018 to February 2021 were retrospectively evaluated. Radiomic features pre-PRRT evaluated using a custom MIM Software® LesionID workflow. Conventional metrics burden, such expression and volume, calculated addition indices for each lesion (intra-lesional) then summarized across all lesions throughout body (inter-lesional). Endpoints included post-PRRT 24-month time progression (TTP) overall survival (OS). Cox regression models used assess ability imaging factors TTP OS. LASSO-penalized was build multivariable model outcome. Results Eighty with mean age 65.1 years included, most (71.3%) completing 4 cycles Median 19.1 months, OS 60 months 50%. A large degree variability between evidenced related expression. On analysis, total liver-corrected SUVmean selected TTP. The not able significantly predict (C-statistic = 0.58, 95% CI 0.50–0.62). Total skewness resulting death 0.62, 0.53–0.67), but limited, by low C-statistic. Conclusions Our exploratory analysis provides preliminary results showing that inter- intra-tumor pretreatment images therapy. However, prospective evaluation larger cohort is needed further whether comprehensive characterization within patient can help guide decisions.
Язык: Английский
Процитировано
1Clinical and Translational Imaging, Год журнала: 2022, Номер 11(4), С. 321 - 328
Опубликована: Дек. 12, 2022
Язык: Английский
Процитировано
4Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 49 - 58
Опубликована: Янв. 1, 2024
Язык: Английский
Процитировано
0