Deep neural network analysis employing diffusion basis spectrum imaging metrics as classifiers improves prostate cancer detection and grading
Ye, Z.; Yang, Q.; Lin, J.; Sun, P.; Shao, C.; Yu, Y.; Chen, L.; Zhu, Y.; Wu, A.; George, A.; Song, C.; Yang, R.; Zhan, J.; Gary, S.; Viox, J.; Wang, Z.; Wang, M.; Chen, Y.; Kim, E. H.; Ippolito, J.; Lu, J.; Song, S.-K.
Show abstract
Structural and cellular complexity of prostatic histopathology limits the accuracy of noninvasive detection and grading of prostate cancer (PCa). We addressed this limitation by employing a novel diffusion basis spectrum imaging (DBSI) to derive structurally-specific diffusion fingerprints reflecting various underlying prostatic structural and cellular components. We further developed diffusion histology imaging (DHI) by combining DBSI-derived structural fingerprints with a deep neural network (DNN) algorithm to more accurately classify different histopathological features and predict tumor grade in PCa. We examined 243 patients suspected with PCa using in vivo DBSI. The in vivo DBSI-derived diffusion metrics detected coexisting prostatic pathologies distinguishing inflammation, PCa, and benign prostatic hyperplasia. DHI distinguished PCa from benign peripheral and transition zone tissues with over 95% sensitivity and specificity. DHI also demonstrated over 90% sensitivity and specificity for Gleason score noninvasively. We present DHI as a novel diagnostic tool capable of noninvasive detection and grading of PCa. One sentence summaryDiffusion histology imaging noninvasively and accurately detects and grades prostate cancer.
Matching journals
The top 10 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Real-time, functional intra-operative localization of rat cavernous nerve network using near-infrared cyanine voltage-sensitive dye imaging 91%
- Improving Rectal Tumor Segmentation with Anomaly Fusion Derived from Anatomical Inpainting: A Multicenter Study 91%
- Correcting B0 inhomogeneity-induced distortions in whole-body diffusion MRI of bone metastases 90%
Similar papers in this journal
- Manganese-enhanced MRI (MEMRI) in breast and prostate cancers: exploring the potential role of calcium receptors. 92%
- Image-localized Biopsy Mapping of Brain Tumor Heterogeneity: A Single-Center Study Protocol 91%
- Glycoprofiling of proteins as prostate cancer biomarkers: a multinational population study 90%
Similar papers in this journal
- Digital Spatial Profiling identifies phospho-JNK as a biomarker for early risk stratification of aggressive prostate cancer 92%
- The expression of PKM1 and PKM2 in developing, benign, and cancerous prostatic tissues 92%
- Deep-Learning-Based Generation of Synthetic High-Resolution MRI from Low-Resolution MRI for Use in Head and Neck Cancer Adaptive Radiotherapy 91%
Similar papers in this journal
Similar papers in this journal
- Human spinal cord activation during filling and emptying of the bladder 90%
- TNF Blockade Reduces Prostatic Hyperplasia and Inflammation while Limiting BPH Diagnosis in Patients with Autoimmune Disease 90%
- Teacher-student collaborated multiple instance learning for pan-cancer PDL1 expression prediction from histopathology slides 88%
"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.