MRI with deep learning–based reconstruction at 1 mm showed higher diagnostic performance than conventional 3-mm slice reconstruction in identifying cavernous sinus invasion by postoperative residua... Background Achieving high-spatial-resolution pituitary MRI is challenging because of the trade-off between image noise and spatial resolution. Deep learning–based MRI reconstruction enables image d...

Purpose: Heterogeneity in glioblastomas is associated with poorer outcomes, and physiologic heterogeneity can be quantified with non-invasive imaging. We developed spatial habitats based on multiparametric physiologic magnetic resonance imaging (MRI) and evaluated associations between temporal changes in these habitats and progression-free survival (PFS) after concurrent chemoradiotherapy (CCRT) in patients with glioblastoma. Experimental Design: Ninety-seven patients with IDH-wild type glioblastoma were enrolled and two serial MRI examinations after CCRT were analyzed. Cerebral blood volumes and apparent diffusion coefficients were grouped using k-means clustering into three spatial habitats. . Associations between temporal changes in spatial habitats and PFS were investigated using Cox proportional hazard modeling. The performance of significant predictors for PFS and overall survival (OS) was measured using a discrete increase of habitat (habitat risk score) in a temporal validation set from a prospective registry (n = 53, ClinicalTrials.gov NCT02619890). The site of progression was matched with the spatiotemporal habitats. Results: Three spatial habitats of hypervascular cellular, hypovascular cellular, and nonviable tissue were identified. A short-term increase in the hypervascular cellular habitat (hazard ratio [HR] 40.0, P=.001) and hypovascular cellular habitat was significantly associated with shorter PFS (HR 3.78, P<.001) after CCRT. Combined with clinical predictors, the habitat risk score showed a C-index of 0.79 for PFS and 0.74 for OS and stratified patients with short-, intermediate-, and long-PFS (P=.016). An increase in the hypovascular cellular habitat predicted tumor-progression sites. Conclusions: Hypovascular cellular habitats derived from multiparametric physiologic MRIs may be useful predictors of clinical outcomes in patients with post-treatment glioblastoma.

Bae SH, et al. J Korean Med Sci. 2014 Jul;29(7):980-984. http://dx.doi.org/10.3346/jkms.2014.29.7.980

Lee DH, et al. Korean J Radiol. 2023 Feb;24:e11. https://doi.org/10.3348/kjr.2022.0492

Background: This study evaluated whether generative artificial intelligence (AI)-based augmentation (GAA) can provide diverse and realistic imaging phenotypes and improve deep learning-based classification of isocitrate dehydrogenase (IDH) type in glioma compared with neuroradiologists. Methods: For model development, 565 patients (346 IDH-wildtype, 219 IDH-mutant) with paired contrast-enhanced T1 and FLAIR MRI scans were collected from tertiary hospitals and The Cancer Imaging Archive. Performance was tested on internal (119, 78 IDH-wildtype, 41 IDH-mutant [IDH1 and 2]) and external test sets (108, 72 IDH-wildtype, 36 IDH-mutant). GAA was developed using a score-based diffusion model and ResNet50 classifier. The optimal GAA was selected in comparison with the null model. Two neuroradiologists (R1, R2) assessed realism, diversity of imaging phenotypes, and predicted IDH mutation. The performance of a classifier trained with optimal GAA was compared with that of neuroradiologists using the area under the receiver operating characteristics curve (AUC). The effect of tumor size and contrast enhancement on GAA performance was tested. Results: Generated images demonstrated realism (Turing’s test: 47.5–50.5%) and diversity indicating IDH type. Optimal GAA was achieved with augmentation with 110 000 generated slices (AUC: 0.938). The classifier trained with optimal GAA demonstrated significantly higher AUC values than neuroradiologists in both the internal (R1, P = .003; R2, P < .001) and external test sets (R1, P < .01; R2, P < .001). GAA with large-sized tumors or predominant enhancement showed comparable performance to optimal GAA (internal test: AUC 0.956 and 0.922; external test: 0.810 and 0.749). Conclusions: The application of generative AI with realistic and diverse images provided better diagnostic performance than neuroradiologists for predicting IDH type in glioma.

BACKGROUND AND PURPOSE: Accurate radiologic prediction of cavernous sinus invasion by pituitary adenoma remains challenging. We aimed to assess whether 1-mm-slice-thickness MRI with deep learning–based reconstruction can better predict cavernous sinus invasion by pituitary adenoma preoperatively and to estimate the depth of invasion and degree of contact in relation to the carotid artery, compared with 3-mm-slice-thickness MRI. MATERIALS AND METHODS: This single-institution, prospective study included 67 consecutive patients (mean age, 53 [SD, 12] years; 28 women), between January and August 2020, who underwent a combined contrast-enhanced T1-weighted imaging protocol of 1-mm-slice-thickness MRI + deep learning–based reconstruction and 3-mm-slice-thickness MRI. An expert neuroradiologist who was blinded to the imaging protocol determined cavernous sinus invasion using the modified Knosp classification on 1-mm-slice-thickness MRI + deep learning–based reconstruction and 3-mm-slice-thickness MRI, respectively. Reference standards were established by the consensus of radiologic, intraoperative, pathologic, and laboratory findings. The primary end point was the diagnostic performance of each imaging protocol, and the secondary end points included depth of invasion and degree of contact in relation to the carotid artery. RESULTS: The diagnostic performance of 1-mm-slice-thickness MRI + deep learning–based reconstruction (area under the curve, 0.79; 95% CI, 0.69 − 0.89) in predicting cavernous sinus invasion by pituitary adenoma was higher than that of 3-mm-slice-thickness MRI (area under the curve, 0.61; 95% CI, 0.52–0.70; P < .001). One-millimeter-slice-thickness MRI + deep learning–based reconstruction demonstrated greater depth of invasion by pituitary adenomas from the medial intercarotid line than 3-mm-slice-thickness MRI (4.07 versus 3.12 mm, P < .001). A higher proportion of cases were in a greater degree of contact with the intracavernous ICA with 1-mm-slice-thickness MRI + deep learning–based reconstruction than with 3-mm-slice-thickness MRI (total encasement, 37.3% versus 13.4%, P < .001; >270°, 38.8% versus 16.4%, P < .001). CONCLUSIONS: Compared with 3-mm-slice-thickness MRI, 1-mm-slice-thickness MRI + deep learning–based reconstruction showed a higher diagnostic performance in preoperatively predicting cavernous sinus invasion by pituitary adenomas and demonstrated a greater depth and degree of contact in relation to the carotid artery.