Utility of quantitative measurement of T2 using Restriction Spectrum Imaging for detection of clinically significant prostate cancer
Rojo Domingo, M.; Conlin, C. C.; Karunamuni, R. A.; Ollison, C.; Baxter, M. T.; Kallis, K.; Do, D. D.; Song, Y.; Kuperman, J. M.; Shabaik, A. S.; Hahn, M. E.; Murphy, P. M.; Rakow-Penner, R. R.; Dale, A. M.; Seibert, T. M.
Show abstract
BackgroundThe Restriction Spectrum Imaging restriction score (RSIrs) has demonstrated higher diagnostic accuracy for clinically significant prostate cancer (csPCa) than conventional DWI. Both diffusion and T2 properties of prostate tissue inform the RSI signal, and studies have shown that each may be valuable for csPCa discrimination. PurposeTo determine whether prostate T2 varies across RSI compartments and in the presence of csPCa, and to evaluate whether consideration of compartmental T2 (cT2) improves csPCa detection over RSIrs alone. Study TypeRetrospective. PopulationTwo cohorts (46 and 195 patients) scanned for csPCa. Field Strength/Sequence3T multi-b-value DWI acquired at multiple TEs. AssessmentcT2 values were computed from multi-TE RSI data and compared between RSI model compartments. csPCa detection was compared between RSIrs and a logistic regression model (LRM) for predicting the probability of csPCa using cT2 in combination with RSI measurements. Statistical TestsTwo-sample t-tests (=0.05) were used to compare cT2 values between compartments and between patients with and without csPCa. Area under the receiver operating characteristic curve (AUC) was used to evaluate csPCa detection performance. ResultsIn both cohorts, T2 differed (p<0.05) across all RSI compartments (C1, C2, C3, C4). Voxel-level data from cohort 1 showed that T2 differed between normal and cancerous tissue in C1, C2, C3 (p<0.001). Whole-prostate T2 differed between patients with and without csPCa in C3 (p=0.02). In cohort 2, whole-prostate T2 differed in C1 (p=0.01), C3 (p=0.01), and C4 (p<0.01). Consideration of cT2 improved csPCa discrimination compared to diffusion alone, but not compared to RSIrs [cohort 1: 0.80 vs 0.70 (diffusion) and 0.80 (RSIrs), cohort 2: 0.72 vs 0.65 (diffusion) and 0.72 (RSIrs)]. Data ConclusionSignificant differences in cT2 were observed between normal and cancerous prostatic tissue. With our data, however, consideration of cT2 did not significantly improve cancer detection performance over RSIrs alone.
Matching journals
The top 9 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Improving Rectal Tumor Segmentation with Anomaly Fusion Derived from Anatomical Inpainting: A Multicenter Study 92%
- Correcting B0 inhomogeneity-induced distortions in whole-body diffusion MRI of bone metastases 91%
- PathProfiler: Automated Quality Assessment of Retrospective Histopathology Whole-Slide Image Cohorts by Artificial Intelligence, A Case Study for Prostate Cancer Research 90%
Similar papers in this journal
- Improved characterization of diffusion in normal and cancerous prostate tissue through optimization of the restriction spectrum imaging signal model 92%
- Test-retest reproducibility of in vivo magnetization transfer ratio and saturation index in mice at 9.4 Tesla 89%
- T2-weighted T1 mapping and automated segmentation of CSF: Assessment of solute gradients in the healthy brain 89%
Similar papers in this journal
- Association Between Artificial Intelligence-Derived Tumor Volume and Oncologic Outcomes for Localized Prostate Cancer Treated with Radiation Therapy 94%
- Use of focal radiotherapy boost for prostate cancer and perceived barriers toward its implementation: a survey 90%
- Clinical Impact of Contouring Variability for Prostate Cancer Tumor Boost 90%
Similar papers in this journal
- Diffusion Histology Imaging Combining Diffusion Basis Spectrum Imaging (DBSI) and Machine Learning Improves Detection and Classification of Glioblastoma Pathology 89%
- PSMA+ Extracellular Vesicles are a Biomarker for SABR in Oligorecurrent Prostate Cancer Analysis from the STOMP-like and ORIOLE trial cohorts 88%
- Discrimination of breast cancer from healthy breast tissues using a three-component diffusion-weighted MRI model 88%
Similar papers in this journal
- Deep-Learning-Based Generation of Synthetic High-Resolution MRI from Low-Resolution MRI for Use in Head and Neck Cancer Adaptive Radiotherapy 91%
- Segmentation stability of human head and neck medical images for radiotherapy applications under de-identification conditions: benchmarking for data sharing and artificial intelligence use-cases 90%
- Digital Spatial Profiling identifies phospho-JNK as a biomarker for early risk stratification of aggressive prostate cancer 89%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.