Search

Purpose To evaluate the diagnostic performance of deep learning using the Residual Networks 50 (ResNet50) neural network constructed from different segmentations for distinguishing malignant and benign non-mass enhancement (NME) on breast magnetic resonance imaging (MRI) and conduct a comparison with radiologists with various levels of experience. Materials and methods A total of 84 consecutive patients with 86 lesions (51 malignant, 35 benign) presenting NME on breast MRI were analyzed. Three radiologists with different levels of experience evaluated all examinations, based on the Breast Imaging-Reporting and Data System (BI-RADS) lexicon and categorization. For the deep learning method, one expert radiologist performed lesion annotation manually using the early phase of dynamic contrast-enhanced (DCE) MRI. Two segmentation methods were applied: a precise segmentation was carefully set to include only the enhancing area, and a rough segmentation covered the whole enhancing region, including the intervenient non-enhancing area. ResNet50 was implemented using the DCE MRI input. The diagnostic performance of the radiologists’ readings and deep learning were then compared using receiver operating curve analysis. Results The ResNet50 model from precise segmentation achieved diagnostic accuracy equivalent [area under the curve (AUC) = 0.91, 95% confidence interval (CI) 0.90, 0.93] to that of a highly experienced radiologist (AUC = 0.89, 95% CI 0.81, 0.96; p = 0.45). Even the model from rough segmentation showed diagnostic performance equivalent to a board-certified radiologist (AUC = 0.80, 95% CI 0.78, 0.82 vs. AUC = 0.79, 95% CI 0.70, 0.89, respectively). Both ResNet50 models from the precise and rough segmentation exceeded the diagnostic accuracy of a radiology resident (AUC = 0.64, 95% CI 0.52, 0.76). Conclusion These findings suggest that the deep learning model from ResNet50 has the potential to ensure accuracy in the diagnosis of NME on breast MRI.

Purpose The aim of this study is to compare lateral ventricular cerebrospinal fluid (CSF) temperature of the patients with Alzheimer's disease (AD), mild cognitive impairment (MCI), and healthy subjects (HS) using diffusion-weighted imaging (DWI)-based magnetic resonance (MR) thermometry. Methods Seventy-two patients (37 AD, 19 MCI, 16 HS) who underwent 3-T MR examination from September 2018 to August 2019 were included in this study. Smoking habits, education level, disease duration, and comorbidity status were recorded. Patients were assessed using Mini-Mental State Examination (MMSE) and the Clinical Dementia Rating (CDR) score. Brain temperatures were measured using DWI-based MR thermometry. Group comparisons of brain temperature were performed using the Pearson chi-square, Mann-Whitney, and Kruskal-Wallis tests. Further analysis was performed using the post hoc Bonferroni test. Receiver operating characteristic (ROC) analysis was also used. Results A CDR score of 0.5, 1, and 2 was 2 (5.4%), 14 (37.8%), and 21 (56.8%) in AD, respectively. The median MMSE score had significant differences among groups and also in pairwise comparisons. The median CSF temperature (degrees C) values showed statistically significant difference among groups (HS: 38.5 degrees C, MCI: 38.17 degrees C, AD: 38.0 degrees C). The post hoc Mann-Whitney U test indicated a significant difference between AD patients and HS (p = 0.009). There were no significant CSF temperature differences in other pairwise comparisons. Conclusion Lower CSF temperatures were observed in AD patients than in HS, probably due to decreased brain metabolism in AD. DWI-based MR thermometry as a noninvasive imaging method enabling the measurement of CSF temperatures may contribute to the diagnosis of AD.

Painful ganglion cysts that develop in the hallux and finger usually enlarge progressively to the peripheral direction. Simple resection of satellite ganglion cyst alone has been reported to cause a high rate of recurrence and treatment is often very difficult. The purpose of this study is to evaluate the appropriate surgical treatment for painful satellite ganglion cysts in the hallux and finger and discuss the origin of the ganglion cysts in cases treated surgically at our hospital. We reviewed five cases (three males and two females, ages 55–87 years), three of which occurred in the hallux and two in the finger. In all cases, the preoperative magnetic resonance image showed a large fluid of the flexor tendon sheath. And also, joint effusion was found in the metatarsophalangeal joint and the proximal interphalangeal joint. The first case of the hallux ganglion underwent simple excision of the cyst and had recurrences three times. In the other four cases, the additional synovectomy of the metatarsophalangeal joint and the proximal interphalangeal joint was performed along with ganglion cyst excision. These cases had no recurrence up to 1 year after operation. Recently, there have been reports that tendon sheath ganglions are connected to the ankle, wrist, hallux, and phalangeal joints. Although there are a few cases in our department, satellite ganglion cyst of the hallux and finger possibly originates from adjacent joints. Additional synovectomy of the affected joint should be performed for the excision of satellite ganglion cyst to prevent recurrence.

To evaluate the relationship of the signature index (S-index), an advanced diffusion MRI marker, and the immunohistochemical (IHC) status, proliferation rate, and molecular subtype of invasive breast cancers.A retrospective study of patients with invasive carcinoma was conducted between 2017 and 2021. All patients underwent dynamic contrast-enhanced MRI and DWI using a 3-T system. For DWI, three b values (0, 200, and 1500 s/mmThe study included 153 patients (mean age, 60 ± 13 years) with 160 invasive carcinomas. S-index values were significantly higher in estrogen receptor-positive (mean, p = .005; 85The DWI S-index showed significantly higher values in invasive carcinomas with immunohistochemical status associated with good prognosis, suggesting its usefulness as a noninvasive imaging biomarker to estimate IHC status and orient treatment.• The signature index, an advanced diffusion MRI marker, showed good discrimination of immunohistochemical status in invasive breast carcinomas. • The signature index has the potential to differentiate noninvasively invasive breast carcinoma subtypes and appears as an imaging biomarker of prognostic factors and molecular phenotypes.

BACKGROUND Few studies have examined the effect of low-grade intraventricular hemorrhage (IVH) on the white matter in the cerebellum and its association with neurodevelopment. We evaluated cerebellar white matter at term-equivalent age (TEA) in preterm infants with low-grade IVH. Furthermore, we assessed neurodevelopmental outcomes at 3 years of age to examine the influence of low-grade IVH on neurodevelopment. METHODS Thirteen infants with low-grade IVH and 26 without IVH, born at

Diffusion-weighted imaging (DWI) thermometry is a magnetic resonance-based imaging tool that allows the noninvasive measurement of brain core temperature. Although only applicable to cerebrospinal fluid, it is thought to be potentially useful in assessing the thermal pathophysiology of the brain in both patients and healthy subjects. The objective of this article is to provide a concise but thorough review of the basic physical principles and the principal applications of DWI thermometry as a potential method to elucidate the pathophysiology of several brain diseases and neurologic syndromes.

We demonstrate magnetometry of cultured neurons on a polymeric film using a superconducting flux qubit that works as a sensitive magnetometer in a microscale area. The neurons are cultured in Fe$^{3+}$ rich medium to increase magnetization signal generated by the electron spins originating from the ions. The magnetometry is performed by insulating the qubit device from the laden neurons with the polymeric film while keeping the distance between them around several micrometers. By changing temperature (12.5 - 200 mK) and a magnetic field (2.5 - 12.5 mT), we observe a clear magnetization signal from the neurons that is well above the control magnetometry of the polymeric film itself. From electron spin resonance (ESR) spectrum measured at 10 K, the magnetization signal is identified to originate from electron spins of iron ions in neurons. This technique to detect a bio-spin system can be extended to achieve ESR spectroscopy at the single-cell level, which will give the spectroscopic fingerprint of cells.

Objectives Parkinson’s disease (PD) is known to be related to various factors, including neuroinflammation, increased oxidative stress, and brain temperature alteration. We aimed to evaluate the correlation between these factors using diffusion-weighted imaging (DWI) thermometry and blood tests of systemic inflammation. Methods From July 2012 to Jun 2017, 103 patients with PD (44 men and 59 women; mean age, 60.43 ± 9.12 years) and 106 sex- and age-matched healthy volunteers (48 men and 58 women; mean age, 58.16 ± 8.45 years) retrospectively underwent magnetic resonance DWI thermometry to estimate brain intraventricular temperature (Tv). Subjects were divided into three subgroups in light of their ages. The tested inflammatory markers included plasma nuclear DNA, mitochondrial DNA, apoptotic leukocytes, and serum adhesion molecules. The correlations among the Tv values, clinical severity, and systemic inflammatory markers were then calculated. Results The PD patients did not show a natural trend of decline in Tv with age. Comparisons among the different age groups revealed that the younger PD subjects had significantly lower Tv values than the younger controls, but the older subjects had no significant group differences. Overall, the PD patients exhibited lower Tv values than the controls, as well as increased oxidative stress. The brain temperature showed positive correlations with inflammatory markers, including plasma nuclear DNA and L-selectin levels, in all the subjects. Conclusions Possible pathophysiological correlations between systemic inflammation and brain temperature were indicated by the results of this study, a finding which may aid us in investigating the underlying pathogenesis of PD.

Radiology report generation through chest radiography interpretation is a time-consuming task that involves the interpretation of images by expert radiologists. It is common for fatigue-induced diagnostic error to occur, and especially difficult in areas of the world where radiologists are not available or lack diagnostic expertise. In this research, we proposed a multi-objective deep learning model called CT2Rep (Computed Tomography to Report) for generating lung radiology reports by extracting semantic features from lung CT scans. A total of 458 CT scans were used in this research, from which 107 radiomics features and 6 slices of segmentation related nodule features were extracted for the input of our model. The CT2Rep can simultaneously predict position, margin, and texture, which are three important indicators of lung cancer, and achieves remarkable performance with an F1-score of 87.29%. We conducted a satisfaction survey for estimating the practicality of CT2Rep, and the results show that 95% of the reports received satisfactory ratings. The results demonstrate the great potential in this model for the production of robust and reliable quantitative lung diagnosis reports. Medical personnel can obtain important indicators simply by providing the lung CT scan to the system, which can bring about the widespread application of the proposed framework.