A three-dimensional convolutional neural network model was developed to classify the severity of chronic kidney disease (CKD) using magnetic resonance imaging (MRI) Dixon-based T1-weighted in-phase (IP)/opposed-phase (OP)/water-only (WO) imaging. Seventy-three patients with severe renal dysfunction (estimated glomerular filtration rate [eGFR] < 30 mL/min/1.73 m², CKD stage G4–5); 172 with moderate renal dysfunction (30 ≤ eGFR < 60 mL/min/1.73 m², CKD stage G3a/b); and 76 with mild renal dysfunction (eGFR ≥ 60 mL/min/1.73 m², CKD stage G1–2) participated in this study. The model was applied to the right, left, and both kidneys, as well as to each imaging method (T1-weighted IP/OP/WO images). The best performance was obtained when using bilateral kidneys and IP images, with an accuracy of 0.862 ± 0.036. The overall accuracy was better for the bilateral kidney models than for the unilateral kidney models. Our deep learning approach using kidney MRI can be applied to classify patients with CKD based on the severity of kidney disease.

To evaluate the clinical utility of intravenous gadolinium-enhanced heavily T2-weighted 3D fluid-attenuated inversion recovery (HT2-FLAIR) imaging for identifying spinal cerebrospinal fluid (CSF) leaks in patients with spontaneous intracranial hypotension (SIH). Patients with SIH underwent MR myelography and post-contrast HT2-FLAIR imaging after an intravenous gadolinium injection. Two types of CSF leaks (epidural fluid collection and CSF leaks around the nerve root sleeve) at each vertebral level were compared between the 2 sequences. The total numbers of CSF leaks and vertebral levels involved were recorded for the whole spine. The sequence that was superior for the overall visualization of epidural and paraspinal fluid collection was then selected. Nine patients with SIH were included in the present study. HT2-FLAIR imaging was equivalent or superior to MR myelography at each level for detecting the 2 types of CSF leaks. In the 2 types of CSF leaks, the total numbers of CSF leaks and levels involved were higher on HT2-FLAIR images than on MR myelography, while no significant difference was observed for CSF leaks around the nerve root sleeve. In all 9 patients, HT2-FLAIR imaging was superior to MR myelography for the overall visualization of epidural and paraspinal fluid collection. Intravenous gadolinium-enhanced HT2-FLAIR imaging was superior to MR myelography for the visualization of CSF leaks in patients with SIH. This method can be useful for identifying spinal CSF leaks.

Newly developed simultaneous multi-slice (SMS) scans are now being introduced as a clinical application in neuroimaging. We examined the feasibility of SMS scans for joint imaging. The purpose of the present study was to prospectively compare the resolution and specific absorption rate (SAR) obtained using SMS to those of conventional methods in hip joint magnetic resonance imaging (MRI) and establish whether imaging times may be reduced using SMS in 3 T MRI and if image quality is affected. Twenty-one patients (4 men and 17 women, average age, 51.5 years, range: 22 to 76 years) with hip pain underwent MR examinations of the unilateral hip joint. Three board-certified radiologists independently and blindly evaluated the images obtained with and without SMS using window and level settings and magnification according to personal preferences. Individual SAR values were measured for each protocol. A Wilcoxon signed-rank test and a t-statistic test were used for statistical analyses. Signal-to-noise ratio (SNR) was also compared using a phantom. SMS imaging maintained equivalent image quality to conventional imaging for evaluating the morphology of the hip joint, and also reduced imaging times by approximately 40%. SMS images had significantly higher SAR values than conventional images. The rate of difference (SMS/conventional) in SNR ranged between 80 and 111%. Based on its significantly lower acquisition times and the maintenance of similar image quality to conventional imaging, SMS may be applied to morphological evaluations of hip joint disorders without significantly increasing SAR.

This study performed three-dimensional (3D) magnetic resonance imaging (MRI)-based statistical shape analysis (SSA) by comparing patellofemoral instability (PFI) and normal femur models, and developed a machine learning (ML)-based prediction model. Twenty (19 patients) and 31 MRI scans (30 patients) of femurs with PFI and normal femurs, respectively, were used. Bone and cartilage segmentation of the distal femurs was performed and subsequently converted into 3D reconstructed models. The pointwise distance map showed anterior elevation of the trochlea, particularly at the central floor of the proximal trochlea, in the PFI models compared with the normal models. Principal component analysis examined shape variations in the PFI group, and several principal components exhibited shape variations in the trochlear floor and intercondylar width. Multivariate analysis showed that these shape components were significantly correlated with the PFI/non-PFI distinction after adjusting for age and sex. Our ML-based prediction model for PFI achieved a strong predictive performance with an accuracy of 0.909 ± 0.015, and an area under the curve of 0.939 ± 0.009 when using a support vector machine with a linear kernel. This study demonstrated that 3D MRI-based SSA can realistically visualize statistical results on surface models and may facilitate the understanding of complex shape features.

Background: Non-contrast FLAIR revealed increased signal within the inner ear in patients with vestibular schwannoma, which is generally assumed to occur in the perilymph; however, the majority of previous studies did not differentiate between the endolymph and perilymph. Therefore, endolymph signal changes have not yet been investigated in detail. The purpose of the present study was three-fold: (1) to assess perilymph signal changes in patients with vestibular schwannoma on heavily T2-weighted (T2W) 3D FLAIR, also termed positive perilymphatic images (PPI), (2) to evaluate signal and morphological changes in the endolymph on PPI, and (3) to establish whether vertigo correlates with the signal intensity ratios (SIR) of the vestibular perilymph or vestibular endolymphatic hydrops. Methods: Forty-two patients with unilateral vestibular schwannoma were retrospectively recruited. We semi-quantitatively and qualitatively evaluated the perilymph signal intensity on the affected and unaffected sides. We also quantitatively examined the signal intensity of the vestibular perilymph and assessed the relationship between vertigo and the SIR of the vestibular perilymph on the affected side. We semi-quantitatively or qualitatively evaluated the endolymph, and investigated whether vestibular hydrops correlated with vertigo. Results: The perilymph on the affected side showed abnormal signal more frequently (signal intensity grade: overall mean 1.45 vs. 0.02; comparison of signal intensity: overall mean 36 vs. 0 cases) and in more parts (the entire inner ear vs. the basal turn of the cochlea and vestibule) than that on the unaffected side. No significant difference was observed in the SIR of the vestibular perilymph with and without vertigo (5.54 vs. 5.51, p = 0.18). The endolymph of the vestibule and semicircular canals showed the following characteristic features: no visualization (n = 4), signal change (n = 1), or vestibular hydrops (n = 10). A correlation was not observed between vestibular hydrops and vertigo (p = 1.000). Conclusions: PPI may provide useful information on signal and morphological changes in the endolymph of patients with vestibular schwannoma. Further research is warranted to clarify the relationship between vertigo and the MR features of the inner ear.