Siqi Cai^1,2, Zhifeng Shi³, Chunxiang Jiang^1,2, Shihui Zhou^1,2, and Lijuan Zhang*^1
1Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, ²University of Chinese Academy of Sciences, Beijing, China, ³Huashan Hospital of Fudan University, Shanghai, China

Thalamic plasticity plays an important role in the functional remodeling in brains with glioma. This study investigated the frequency-specific functional alteration of the thalami in the context of gliomas with discrete wavelet transformation and wavelet-energy. Brains with glioma in the left hemisphere featured with amplified inter-thalamus energy deviation for BOLD signals at 15.63-250mHz irrespective of the malignant grades, as compared with healthy controls. Similar analysis revealed no energy changes for the right brain gliomas. These findings suggest a differential thalamic plasticity that is specific to the tumor laterality, and may leverage innovation in the treatment of glioma involving neuromodulation.

Yida Wang¹, Naying He², Chenglong Wang¹, Yan Li², Zhijia Jin², Xiance Zhao³, Ewart Mark Haacke^2,4, Fuhua Yan², and Guang Yang^1
1Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China, ²Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, ³Philips Healthcare, Shanghai, China, ⁴Department of Biomedical Engineering, Wayne State University, Detroit, MI, United States

We proposed an automatic cascaded framework based on deep learning to segment deep brain nuclei and distinguish Parkinson’s disease from normal controls using quantitative susceptibility mapping (QSM) images. A 3D CA-Net model integrating channel attention, spatial attention and scale attention module was utilized to segment 5 brain nuclei from QSM and T1W data. Then, the QSM images and the segmented brain nuclei ROIs were fed into the SE-ResNeXt50 with anatomical attention mechanism to get the predicted PD probability. The proposed method provided good interpretability and achieved AUC values of 0.97 and 0.90 on training and testing cohort, respectively.

Su Dongning^1,2, Zhang Zhe^2,3, Zhu Wanlin^2,3, Sui Binbin^2,3, Zhang Yingkui^2,3, Bi Jingfeng^2,3, Kong Qingle ⁴, Zhang Zhijin^1,2, Gan Yawen^1,2, Jin Jianing^1,2, Wang Xuemei^1,2, Wang Zhan^1,2, Wang Yongjun ^1,2,3, Wu Tao^1,2, Jing Jing^1,2,3, and Feng Tao^1,2
1Department of Neurology, Beijing Tiantan Hospital,Captital Medical University, Beijing, China, ²China National Clinical Research Center for Neurological Diseases, Beijing, China, ³Tiantan Neuroimaging Center of Excellence, Beijing Tiantan Hospital, Capital Medical University, Beijing, China, ⁴MR Collaboration, Siemens Healthineers Ltd., Beijing, China

Ultra-high field MRI enables visualization of anatomical alterations and quantification of iron deposition in the subcortical nucleus. To investigate whether 7T MRI could provide the diagnostic opportunity for Parkinson-plus syndromes, we recruited PD, MSA-P, MSA-C and PSP patients who underwent multi-modal scans including 3D MPRAGE and 3D multi-echo T2*-weighted MRI. All six patients with MSA-C showed visible nigrosome-1. The distances from SN to RN were significantly shorter in PSP than PD. The R2* values in SN, RN and putamen were significantly higher in MSA-P than PD. Our pilot study provided first-of-its-kind evidence that 7T MRI could help diagnose Parkinson-plus syndromes.

Weiwei Zhao¹, Xi Wu², Chunhui Yang², Luguang Chen³, Yiqing Qiu², Chao Ma³, Gaiying Li¹, Yang Song⁴, Yi Wang⁵, and Jianqi Li^1
1East Chian Normal University, Shanghai, China, ²Department of Neurosurgery, Changhai Hospital, Shanghai, China, ³Department of Radiology, Changhai Hospital, Shanghai, China, ⁴MR Scientific Marketing, Siemens Healthineers, Shanghai, China, ⁵Department of Radiology, Weill Medical College of Cornell University, New York, NY, United States

The objective of this study was to investigate the relationship between motor outcomes of subthalamic nuclei deep brain stimulation (STN-DBS) and spatial distribution of iron in the deep gray nuclei for patients with Parkinson's disease (PD). The first and second texture parameters were obtained from the preoperative QSM of 40 PD patients. Significant correlations were found between motor improvement after DBS and QSM texture parameters in substantia nigra (SN) and dentate nucleus (DN). Linear regression showed that second-order texture parameter angular second moment in SN (r = -0.449, p = 0.004) had the highest correlation with the STN-DBS motor outcomes.

Septian Hartono^1,2, Tiet Gan Ong³, Weiling Lee⁴, Zi Cheng Kuek⁴, Celeste Chen¹, Kuan Jin Lee⁵, Jongho Lee⁶, Eng King Tan^1,2, and Ling Ling Chan^2,4
1National Neuroscience Institute, Singapore, Singapore, ²Duke-NUS Graduate Medical School, Singapore, Singapore, ³Nanyang Technological University, Singapore, Singapore, ⁴Singapore General Hospital, Singapore, Singapore, ⁵Institue of Bioengineering and Bioimaging, Singapore, Singapore, ⁶Seoul National University, Seoul, Korea, Republic of

We found associations between tract-specific myelin water fraction and cognitive decline, and motor dysfunction in patients with Parkinson’s disease (PD). Myelin water fractions (MWF) in the anterior corona radiata and posterior thalamic radiations were associated with cognitive decline in PD whilst those in the superior and posterior corona radiata, posterior limb of internal capsule and corticospinal tract were associated with moderate to severe motor dysfunction in PD. Our results suggest that MWF has potential utility as an objective imaging biomarker to monitor longitudinal changes in target WM tracts relevant to motor and cognitive dysfunction in PD neurodegeneration.

Yao-Chia Shih^1,2,3, Pohchoo Seow¹, Hartono Septian^2,4, Welton Thomas^2,4, Weiling Lee¹, Aeden Zi Cheng Kuek¹, Say Lee Chong¹, Samuel Yong Ern Ng⁴, Nicole Shuang Yu Chia⁴, Eng-King Tan⁴, Louis CS Tan⁴, and Ling-Ling Chan^1,2
1Department of Diagnostic Radiology, Singapore General Hospital, Singapore, Singapore, ²Duke-NUS Medical School, Singapore, Singapore, ³Graduate Institute of Medicine, Yuan Ze University, Taoyuan, Taiwan, ⁴Department of Neurology, National Neuroscience Institute – SGH Campus, Singapore, Singapore

Parkinson’s disease (PD) is a common neurodegenerative disease. Besides motor deficits, autonomic and cognitive dysfunctions might be caused by global abnormalities beyond the dopaminergic system. We performed resting-state fMRI to measure functional connectivity between the central autonomic network and eight other critical resting networks in both early PD and healthy control groups. Using general linear models, we identified functional impairments in the inter- and intra-network connections in relation to various brain functions. These findings could shed light on the neural mechanisms behind the complex clinical manifestations evident even in the early stages of PD.

Haiying Lv¹, Qing Li², and Yong Lu^3
1Shanghai Jiao Tong University School of Medicine, Shanghai, China, ²MR Collaborations, Siemens Healthineers Ltd., Shanghai, China, ³Ruijin Hospital / Lu Wan Branch Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China

MRI imaging with Neuromelanin contrast has been thrived in recent years. We adopted this method again to explore the LC contrast differences in multiple system atrophy (MSA). Three methods for quantifying LC contrast were used, which all had the ability to reproduce the results that LC contrast decreases in MSA. We further did subgroup analyses of MSA with orthostatic hypotension (MSA-OH) and MSA without orthostatic hypotension (MSA-nonOH). Interestingly, although LC contrast of the two subgroups were not significantly different, adverse correlations between LC contrast and autonomic dysfunction severity were found in MSA-OH and MSA-nonOH subgroups.

Takahiro Sanada¹, Shota Yamamoto¹, Hirotaka Sato^1,2, Mio Sakai³, Masato Saito¹, Nobuyuki Mitsui¹, Satoru Hiroshima¹, Ryogo Anei⁴, Yonehiro Kanemura⁵, Katsuyuki Nakanishi³, Haruhiko Kishima⁶, and Manabu Kinoshita^1
1Neurosurgery, Asahikawa Medical University, Asahikawa, Japan, ²Neurosurgery, Japanese Red Cross Kitami Hospital, Asahikawa, Japan, ³Diagnostic and Interventional Radiology, Osaka International Cancer Institute, Osaka, Japan, ⁴Neurosurgery, Moriyama Hospital, Asahikawa, Japan, ⁵Biomedical Research and Innovation, National Hospital Organization Osaka National Hospital, Osaka, Japan, ⁶Neurosurgery, Osaka University Graduate School of Medicine, Suita, Japan

The authors investigated the correlation of the ratio of T1WI to T2WI signal intensity (rT1/T2) and IDH mutation status in lower-grade gliomas (LrGG). An exploratory study was conducted to determine an ideal cut-off of rT1/T2 for predicting IDH-mutation, followed by a validation study using two different cohorts. Mean rT1/T of IDH-wild type LrGG was higher than IDH-mutant and significantly predicted IDH mutation in the exploratory cohort. The domestic cohort validated this result, but the accuracy in the TCIA / TCGA cohort declined. The cause of this decline should be further investigated. 

Guohua Zhao¹, Huiting Zhang², Eryuan Gao¹, Ankang Gao¹, Xiaoyue Ma¹, Jie Bai¹, Xu Yan², Guang Yang³, Yusong Lin⁴, and Jingliang Cheng^1
1The Department of Magnetic Resonance Imaging, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China, ²Siemens Healthineers, Shanghai, China, ³Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China, ⁴The School of Software, Zhengzhou University, Zhengzhou, China

Differentiation of glioblastomas and solitary brain metastases is clinically crucial for the prescription of patient management and assessment of prognosis. However, indistinguishable signs between two tumors on routine MRI leads to a high misdiagnosis rate. Neurite orientation dispersion and density imaging (NODDI) can not only estimate the intricacy of neurites in vivo but also provide data to illuminate pathology. We developed a series of radiomics models of NODDI parameter maps, routine MRI, combined routine MRI, and combined NODDI parameter maps to compare their performance in the identification of two tumors. Finally, the combined NODDI radiomics model obtained the best performance.

Jixin Luan¹, Jun Chen¹, Hua Fan¹, Di Zhang¹, Peng Wu², and Chuanchen Zhang^1
1Department of Radiology, Liaocheng People's Hospital, Shandong First Medical University & Shandong Academy of Medical Sciences, Liaocheng, China, ²Philips Healthcare, Shanghai, China

In this study, we aimed to explore the deeper molecular mechanism of radiomics features, and uncover new GBM prognostic biomarker. We extracted radiomics features out of 57 patients with MRI data obtained from TCIA database and corresponding clinical and transcription information from TCGA database. WGCNA revealed that the radiomics features are correlated with numerous cancer-associated biological processes, including regulation of vasculature development and so on. HMGA2 as the hub gene was screened by Cytoscape software. There was significant difference in survival analysis between the high and low expression groups, indicating that HMGA2 may be a new prognostic biomarker for GBM.

Yohan Jun*¹, Yae Won Park*², Hyungseob Shin¹, Yejee Shin¹, Jeong Ryong Lee¹, Kyunghwa Han², Soo Mee Lim³, Seung-Koo Lee², Sung Soo Ahn**², and Dosik Hwang**^1
1Electrical and Electronic Engineering, Yonsei University, Seoul, Korea, Republic of, ²Department of Radiology and Research Institute of Radiological Science and Center for Clinical Imaging Data Science, Yonsei University College of Medicine, Seoul, Korea, Republic of, ³Department of Radiology, Ewha Womans University College of Medicine, Seoul, Korea, Republic of

Preoperative prediction of meningioma grade is important because it influences treatment planning, including surgical resection and stereotactic radiosurgery strategy. The aim of this study was to establish a robust interpretable multiparametric deep learning (DL) model for automatic noninvasive grading of meningiomas along with segmentation. We demonstrated that the interpretable multiparametric DL grading model that combined the T2-weighted and contrast-enhanced T1-weighted images can enable fully automatic grading of meningiomas along with segmentation.

Chang-Le Chen^1,2, Ming-Che Kuo³, Chun-Hwei Tai³, Yung-Chin Hsu⁴, Ruey-Meei Wu³, and Wen-Yih Isaac Tseng^1,5
1Institute of Medical Device and Imaging, National Taiwan University College of Medicine, Taipei, Taiwan, ²Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA, United States, ³Department of Neurology, National Taiwan University Hospital, Taipei, Taiwan, ⁴AcroViz Technology Inc., Taipei, Taiwan, ⁵Molecular Imaging Center, National Taiwan University, Taipei, Taiwan

By applying spatial normative models based on a population-based cohort, we can quantify the microstructural deviation of white matter in patients with Parkinson’s disease (PD) and multiple system atrophy (MSA). We found that microstructural deviation of the frontal aslant tracts, cerebrospinal tracts, and parts of corpus callosum was more profound in MSA compared to that in PD. Also, the uncinate fasciculi were the common degenerative marker in these two diseases. Moreover, the microstructural deviation of WM tracts may reflect the association with the disease-related daily functional deficit in MSA and the timing of disease onset in PD.

Chih-Chien Tsai^1,2, Ting-Wei Liao³, Yih-Ru Wu^3,4, Yi-Ming Wu⁵, and Jiun-Jie Wang^2,6,7
1Department of Medical imaging and Radiological Science, Chang Gung University, Taoyuan City, Taiwan, ²Healthy Aging Research Center, Chang Gung University, Taoyuan City, Taiwan, ³Department of Neurology, Chang Gung Memorial Hospital, Taoyuan City, Taiwan, ⁴College of Medicine, Chang Gung University, Taoyuan City, Taiwan, ⁵Department of f Medical Imaging and Intervention, Chang Gung Memorial Hospital, Taoyuan City, Taiwan, ⁶Department of Medical imaging and Radiological Science, Chang-Gung University, Taoyuan City, Taiwan, ⁷Medical Imaging Research Center, Institute for Radiological Research, Chang Gung University/Chang Gung Memorial Hospital, Taoyuan City, Taiwan

Cognitive decline is a common non-motor symptom in Parkinson's disease (PD) patients, described as mild cognitive impairment (PD-MCI). PD-MCI is identified as a risk for PD with Dementia. Previous studies indicated that PD-MCI patients showed gray matter volume decrease and white matter tract degeneration. We investigate the utility of fixel based analysis as a biomarker for disease progression. The results show a significant degeneration of white matter in PD-MCI patients. Our findings indicated the white matter tract degeneration in PD-MCI patients and shed light on the value of the fiber-specific method to the clinical diagnosis of PD-MCI.

Yasuhiro Fujiwara¹, Shota Ishida², Yuki Matta³, Masayuki Kanamoto³, and Hirohiko Kimura^4
1Department of Life Sciences, Fuculty of Medical Image Sciences, Kumamoto University, Kumamoto, Japan, ²Department of Radiological Technology, Faculty of Medical Sciences, Kyoto College of Medical Science, Kyoto, Japan, ³Radiological Center, University of Fukui Hospital, Fukui, Japan, ⁴Department of Radiology, University of Fukui, Fukui, Japan

An anatomical atlas of the three regions (medial, dorsal, and ventrolateral) of the substantia nigra pars compacta (SNpc) were created based on neuromelanin (NM)-sensitive images, and Proton density, T1, and T2 values for each region were characterized in healthy volunteers. The T1 value of the ventrolateral group was higher than the other SNpc regions, which may be correlated with the NM concentration. The measured relaxation times can be used as baseline values for the SNpc.

Ayaka Yokota¹, Yasuhiro Fujiwara², Tetsuyoshi Hirai³, Nobutaka Sakae⁴, Naokazu Sasagasako⁴, Takahisa Izumi⁵, Kuniharu Ohi⁵, and Hiroyuki Kumazoe^5
1Graduated School of Health Sciences, Kumamoto University, Kumamoto, Japan, ²Department of Medical Image Sciences, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan, ³Faculty of Medicine, Saga University, Saga, Japan, ⁴Department of Neurology, National Hospital Organization Omuta National Hospital, Omuta, Japan, ⁵Department of Radiology, National Hospital Organization Omuta National Hospital, Omuta, Japan

We evaluated the T1 values of the whole and subregions of the substantia nigra (SN) in patients with Parkinson's disease (PD) to identify an imaging biomarker for the early diagnosis of PD. T1 values of the SN of patients with PD were significantly shorter than those of healthy controls and tended to be longer in the lateral part for both groups. T1 values may reflect the pathology of the SN and help assess the degeneration of the SN caused by PD.