Matches in SemOpenAlex for { <https://semopenalex.org/work/W2953441164> ?p ?o ?g. }
- W2953441164 endingPage "251" @default.
- W2953441164 startingPage "243" @default.
- W2953441164 abstract "The volume of pelvic hematoma at CT has been shown to be the strongest independent predictor of major arterial injury requiring angioembolization in trauma victims with pelvic fractures, and also correlates with transfusion requirement and mortality. Measurement of pelvic hematomas (unopacified extraperitoneal blood accumulated from time of injury) using semi-automated seeded region growing is time-consuming and requires trained experts, precluding routine measurement at the point of care. Pelvic hematomas are markedly variable in shape and location, have irregular ill-defined margins, have low contrast with respect to viscera and muscle, and reside within anatomically distorted pelvises. Furthermore, pelvic hematomas occupy a small proportion of the entire volume of a chest, abdomen, and pelvis (C/A/P) trauma CT. The challenges are many, and no automated methods for segmentation and volumetric analysis have been described to date. Traditional approaches using fully convolutional networks result in coarse segmentations and class imbalance with suboptimal convergence. In this study, we implement a modified coarse-to-fine deep learning approach—the Recurrent Saliency Transformation Network (RSTN) for pelvic hematoma volume segmentation. RSTN previously yielded excellent results in pancreas segmentation, where low contrast with adjacent structures, small target volume, variable location, and fine contours are also problematic. We have curated a unique single-institution corpus of 253 C/A/P admission trauma CT studies in patients with bleeding pelvic fractures with manually labeled pelvic hematomas. We hypothesized that RSTN would result in sufficiently high Dice similarity coefficients to facilitate accurate and objective volumetric measurements for outcome prediction (arterial injury requiring angioembolization). Cases were separated into five combinations of training and test sets in an 80/20 split and fivefold cross-validation was performed. Dice scores in the test set were 0.71 (SD ± 0.10) using RSTN, compared to 0.49 (SD ± 0.16) using a baseline Deep Learning Tool Kit (DLTK) reference 3D U-Net architecture. Mean inference segmentation time for RSTN was 0.90 min (± 0.26). Pearson correlation between predicted and manual labels was 0.95 with p < 0.0001. Measurement bias was within 10 mL. AUC of hematoma volumes for predicting need for angioembolization was 0.81 (predicted) versus 0.80 (manual). Qualitatively, predicted labels closely followed hematoma contours and avoided muscle and displaced viscera. Further work will involve validation using a federated dataset and incorporation into a predictive model using multiple segmented features." @default.
- W2953441164 created "2019-07-12" @default.
- W2953441164 creator A5017658063 @default.
- W2953441164 creator A5020505460 @default.
- W2953441164 creator A5049933355 @default.
- W2953441164 creator A5067640436 @default.
- W2953441164 creator A5086706224 @default.
- W2953441164 date "2019-06-07" @default.
- W2953441164 modified "2023-09-28" @default.
- W2953441164 title "Performance of a Deep Learning Algorithm for Automated Segmentation and Quantification of Traumatic Pelvic Hematomas on CT" @default.
- W2953441164 cites W1901129140 @default.
- W2953441164 cites W1903029394 @default.
- W2953441164 cites W1990180140 @default.
- W2953441164 cites W1996243156 @default.
- W2953441164 cites W2001977326 @default.
- W2953441164 cites W2031915343 @default.
- W2953441164 cites W2035136389 @default.
- W2953441164 cites W2063085086 @default.
- W2953441164 cites W2067794541 @default.
- W2953441164 cites W2082198788 @default.
- W2953441164 cites W2102150307 @default.
- W2953441164 cites W2139067100 @default.
- W2953441164 cites W2139805423 @default.
- W2953441164 cites W2146820130 @default.
- W2953441164 cites W2302255633 @default.
- W2953441164 cites W2461474422 @default.
- W2953441164 cites W2463818697 @default.
- W2953441164 cites W2464708700 @default.
- W2953441164 cites W2469107318 @default.
- W2953441164 cites W2526009326 @default.
- W2953441164 cites W2527527874 @default.
- W2953441164 cites W2592929672 @default.
- W2953441164 cites W2604790786 @default.
- W2953441164 cites W2618237340 @default.
- W2953441164 cites W2705158815 @default.
- W2953441164 cites W2738232076 @default.
- W2953441164 cites W2765571304 @default.
- W2953441164 cites W2767236661 @default.
- W2953441164 cites W2771252144 @default.
- W2953441164 cites W2792601183 @default.
- W2953441164 cites W2793594133 @default.
- W2953441164 cites W2962914239 @default.
- W2953441164 cites W3100175091 @default.
- W2953441164 doi "https://doi.org/10.1007/s10278-019-00207-1" @default.
- W2953441164 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7064706" @default.
- W2953441164 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31172331" @default.
- W2953441164 hasPublicationYear "2019" @default.
- W2953441164 type Work @default.
- W2953441164 sameAs 2953441164 @default.
- W2953441164 citedByCount "25" @default.
- W2953441164 countsByYear W29534411642019 @default.
- W2953441164 countsByYear W29534411642020 @default.
- W2953441164 countsByYear W29534411642021 @default.
- W2953441164 countsByYear W29534411642022 @default.
- W2953441164 countsByYear W29534411642023 @default.
- W2953441164 crossrefType "journal-article" @default.
- W2953441164 hasAuthorship W2953441164A5017658063 @default.
- W2953441164 hasAuthorship W2953441164A5020505460 @default.
- W2953441164 hasAuthorship W2953441164A5049933355 @default.
- W2953441164 hasAuthorship W2953441164A5067640436 @default.
- W2953441164 hasAuthorship W2953441164A5086706224 @default.
- W2953441164 hasBestOaLocation W29534411642 @default.
- W2953441164 hasConcept C126838900 @default.
- W2953441164 hasConcept C154945302 @default.
- W2953441164 hasConcept C2777545690 @default.
- W2953441164 hasConcept C2778357063 @default.
- W2953441164 hasConcept C2779662492 @default.
- W2953441164 hasConcept C2779983558 @default.
- W2953441164 hasConcept C41008148 @default.
- W2953441164 hasConcept C71924100 @default.
- W2953441164 hasConcept C89600930 @default.
- W2953441164 hasConceptScore W2953441164C126838900 @default.
- W2953441164 hasConceptScore W2953441164C154945302 @default.
- W2953441164 hasConceptScore W2953441164C2777545690 @default.
- W2953441164 hasConceptScore W2953441164C2778357063 @default.
- W2953441164 hasConceptScore W2953441164C2779662492 @default.
- W2953441164 hasConceptScore W2953441164C2779983558 @default.
- W2953441164 hasConceptScore W2953441164C41008148 @default.
- W2953441164 hasConceptScore W2953441164C71924100 @default.
- W2953441164 hasConceptScore W2953441164C89600930 @default.
- W2953441164 hasIssue "1" @default.
- W2953441164 hasLocation W29534411641 @default.
- W2953441164 hasLocation W29534411642 @default.
- W2953441164 hasLocation W29534411643 @default.
- W2953441164 hasOpenAccess W2953441164 @default.
- W2953441164 hasPrimaryLocation W29534411641 @default.
- W2953441164 hasRelatedWork W2062678678 @default.
- W2953441164 hasRelatedWork W2091515907 @default.
- W2953441164 hasRelatedWork W2265415964 @default.
- W2953441164 hasRelatedWork W2350724852 @default.
- W2953441164 hasRelatedWork W2762462883 @default.
- W2953441164 hasRelatedWork W2763816647 @default.
- W2953441164 hasRelatedWork W2894754340 @default.
- W2953441164 hasRelatedWork W2895323313 @default.
- W2953441164 hasRelatedWork W4282833427 @default.
- W2953441164 hasRelatedWork W4372300963 @default.
- W2953441164 hasVolume "33" @default.
- W2953441164 isParatext "false" @default.