Detecting lumbar lesions in 99m Tc-MDP SPECT by deep learning: Comparison with physicians.

Autor: Petibon Y; Gordon Center of Medical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA.; Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA., Fahey F; Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA.; Division of Nuclear Medicine and Molecular Imaging, Boston Children's Hospital, Boston, Massachusetts, USA., Cao X; Division of Nuclear Medicine and Molecular Imaging, Boston Children's Hospital, Boston, Massachusetts, USA., Levin Z; Gordon Center of Medical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA., Sexton-Stallone B; Division of Nuclear Medicine and Molecular Imaging, Boston Children's Hospital, Boston, Massachusetts, USA., Falone A; Division of Nuclear Medicine and Molecular Imaging, Boston Children's Hospital, Boston, Massachusetts, USA., Zukotynski K; Departments of Medicine and Radiology, McMaster University, Hamilton, Ontario, Canada., Kwatra N; Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA.; Division of Nuclear Medicine and Molecular Imaging, Boston Children's Hospital, Boston, Massachusetts, USA., Lim R; Gordon Center of Medical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA.; Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA., Bar-Sever Z; Institute of Nuclear Medicine, Schneider Children's Medical Center of Israel, Petah Tikva, Israel., Chemli Y; Gordon Center of Medical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA., Treves ST; Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA.; Division of Nuclear Medicine and Molecular Imaging, Brigham and Women's Hospital, Boston, Massachusetts, USA., Fakhri GE; Gordon Center of Medical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA.; Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA., Ouyang J; Gordon Center of Medical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA.; Department of Radiology, Harvard Medical School, Boston, Massachusetts, USA.
Jazyk: angličtina
Zdroj: Medical physics [Med Phys] 2021 Aug; Vol. 48 (8), pp. 4249-4261. Date of Electronic Publication: 2021 Jul 11.
DOI: 10.1002/mp.15033
Abstrakt: Purpose: 99m Tc-MDP single-photon emission computed tomography (SPECT) is an established tool for diagnosing lumbar stress, a common cause of low back pain (LBP) in pediatric patients. However, detection of small stress lesions is complicated by the low quality of SPECT, leading to significant interreader variability. The study objectives were to develop an approach based on a deep convolutional neural network (CNN) for detecting lumbar lesions in 99m Tc-MDP scans and to compare its performance to that of physicians in a localization receiver operating characteristic (LROC) study.
Methods: Sixty-five lesion-absent (LA) 99m Tc-MDP studies performed in pediatric patients for evaluating LBP were retrospectively identified. Projections for an artificial focal lesion were acquired separately by imaging a 99m Tc capillary tube at multiple distances from the collimator. An approach was developed to automatically insert lesions into LA scans to obtain realistic lesion-present (LP) 99m Tc-MDP images while ensuring knowledge of the ground truth. A deep CNN was trained using 2.5D views extracted in LP and LA 99m Tc-MDP image sets. During testing, the CNN was applied in a sliding-window fashion to compute a 3D "heatmap" reporting the probability of a lesion being present at each lumbar location. The algorithm was evaluated using cross-validation on a 99m Tc-MDP test dataset which was also studied by five physicians in a LROC study. LP images in the test set were obtained by incorporating lesions at sites selected by a physician based on clinical likelihood of injury in this population.
Results: The deep learning (DL) system slightly outperformed human observers, achieving an area under the LROC curve (AUC LROC ) of 0.830 (95% confidence interval [CI]: [0.758, 0.924]) compared with 0.785 (95% CI: [0.738, 0.830]) for physicians. The AUC LROC for the DL system was higher than that of two readers (difference in AUC LROC [ΔAUC LROC ] = 0.049 and 0.053) who participated to the study and slightly lower than that of two other readers (ΔAUC LROC  = -0.006 and -0.012). Another reader outperformed DL by a more substantial margin (ΔAUC LROC  = -0.053).
Conclusion: The DL system provides comparable or superior performance than physicians in localizing small 99m Tc-MDP positive lumbar lesions.
(© 2021 American Association of Physicists in Medicine.)
Databáze: MEDLINE
Externí odkaz: