2023
Predicting tumor recurrence on baseline MR imaging in patients with early-stage hepatocellular carcinoma using deep machine learning
Kucukkaya A, Zeevi T, Chai N, Raju R, Haider S, Elbanan M, Petukhova-Greenstein A, Lin M, Onofrey J, Nowak M, Cooper K, Thomas E, Santana J, Gebauer B, Mulligan D, Staib L, Batra R, Chapiro J. Predicting tumor recurrence on baseline MR imaging in patients with early-stage hepatocellular carcinoma using deep machine learning. Scientific Reports 2023, 13: 7579. PMID: 37165035, PMCID: PMC10172370, DOI: 10.1038/s41598-023-34439-7.Peer-Reviewed Original ResearchAAPM Task Group Report 238: 3D C‐arms with volumetric imaging capability*
Supanich M, Siewerdsen J, Fahrig R, Farahani K, Gang G, Helm P, Jans J, Jones K, Koenig T, Kuhls-Gilcrist A, Lin M, Riddell C, Ritschl L, Schafer S, Schueler B, Silver M, Timmer J, Trousset Y, Zhang J. AAPM Task Group Report 238: 3D C‐arms with volumetric imaging capability*. Medical Physics 2023, 50: e904-e945. PMID: 36710257, DOI: 10.1002/mp.16245.Peer-Reviewed Original ResearchConceptsImage qualityGeometric calibrationSystem calibrationPhantom imagesService callsC-arm systemImage reconstructionVolumetric imaging capabilityImage acquisitionGeometric alignmentImage-guided radiation therapyData setsTesting approachCBCT systemC-arm cone beamC-armReconstruction characteristicsImage artifactsValuable dataC-arm CBCT systemImagesIGRT systemTask groupDifferent systemsCT image qualityComparing 3D, 2.5D, and 2D Approaches to Brain Image Auto-Segmentation
Avesta A, Hossain S, Lin M, Aboian M, Krumholz H, Aneja S. Comparing 3D, 2.5D, and 2D Approaches to Brain Image Auto-Segmentation. Bioengineering 2023, 10: 181. PMID: 36829675, PMCID: PMC9952534, DOI: 10.3390/bioengineering10020181.Peer-Reviewed Original ResearchLimited training dataDice scoreComputational memoryTraining dataBrain imagesDeep-learning methodsHigher Dice scoresSegmentation accuracyAuto-segmentation modelComputational speedPerformance metricsOne-sliceAuto-SegmentationBetter performanceConsecutive slicesImagesDeploymentLowest Dice scoresMemoryPerformanceTrainingMetricsModelAccuracyData
2022
DuDoSS: Deep‐learning‐based dual‐domain sinogram synthesis from sparsely sampled projections of cardiac SPECT
Chen X, Zhou B, Xie H, Miao T, Liu H, Holler W, Lin M, Miller EJ, Carson RE, Sinusas AJ, Liu C. DuDoSS: Deep‐learning‐based dual‐domain sinogram synthesis from sparsely sampled projections of cardiac SPECT. Medical Physics 2022, 50: 89-103. PMID: 36048541, PMCID: PMC9868054, DOI: 10.1002/mp.15958.Peer-Reviewed Original ResearchConceptsLow reconstruction accuracySynthetic projectionsAbsolute percent errorImage predictionSPECT image reconstructionImage domainSinogram synthesisGround truthReconstruction accuracyImage reconstructionSinogram domainProjection angleData acquisitionMean square errorFast data acquisitionImagesReconstruction artifactsSPECT imagesSquare error