A comparative analysis of deep learning and hybrid iterative reconstruction algorithms with contrast-enhancement-boost post-processing on the image quality of indirect computed tomography venography of the lower extremities.

PURPOSE: To examine whether there is a significant difference in image quality between the deep learning reconstruction (DLR [AiCE, Advanced Intelligent Clear-IQ Engine]) and hybrid iterative reconstruction (HIR [AIDR 3D, adaptive iterative dose reduction three dimensional]) algorithms on the conventional enhanced and CE-boost (contrast-enhancement-boost) images of indirect computed tomography venography (CTV) of lower extremities. MATERIALS AND METHODS: In this retrospective study, seventy patients who underwent CTV from June 2021 to October 2022 to assess deep vein thrombosis and varicose veins were included. Unenhanced and enhanced images were reconstructed for AIDR 3D and AiCE, AIDR 3D-boost and AiCE-boost images were obtained using subtraction software. Objective and subjective image qualities were assessed, and radiation doses were recorded. RESULTS: The CT values of the inferior vena cava (IVC), femoral vein ( FV), and popliteal vein (PV) in the CE-boost images were approximately 1.3 (1.31-1.36) times higher than in those of the enhanced images. There were no significant differences in mean CT values of IVC, FV, and PV between AIDR 3D and AiCE, AIDR 3D-boost and AiCE-boost images. Noise in AiCE, AiCE-boost images was significantly lower than in AIDR 3D and AIDR 3D-boost images ( P < 0.05). The SNR (signal-to-noise ratio), CNR (contrast-to-noise ratio), and subjective scores of AiCE-boost images were the highest among 4 groups, surpassing AiCE, AIDR 3D, and AIDR 3D-boost images (all P < 0.05). CONCLUSION: In indirect CTV of the lower extremities images, DLR with the CE-boost technique could decrease the image noise and improve the CT values, SNR, CNR, and subjective image scores. AiCE-boost images received the highest subjective image quality score and were more readily accepted by radiologists.

Prognostic Assessment in Patients With Primary Diffuse Large B-Cell Lymphoma of the Central Nervous System Using MRI-Based Radiomics.

BACKGROUND: Primary central nervous system lymphoma (PCNSL) carries a poor prognosis. Radiomics may hold potential value in prognostic assessment. PURPOSE: To develop and validate an MRI-based radiomics model and combine it with clinical factors to assess progression-free survival (PFS) and overall survival (OS) of patients with PCNSL. STUDY TYPE: Retrospective and prospective. POPULATION: Three hundred seventy-nine patients (179 female, 53 ± 7 years) from 2014 to 2022. FIELD STRENGTH/SEQUENCE: T2/fluid-attenuated inversion recovery, contrast-enhanced T1WI and diffusion-weighted echo-planar imaging sequences on 3.0 T. ASSESSMENT: Radiomics features were extracted from enhanced tumor regions on preoperative multi-sequence MRI. Using a least absolute shrinkage and selection operator (LASSO) Cox regression model to select radiomic signatures in training cohort (N = 169). Cox proportional hazards models were constructed for clinical, radiomics, and combined models, with internal (N = 72) and external (N = 32) cohorts validating model performance. STATISTICAL TESTS: Chi-squared, Mann-Whitney, Kaplan-Meier, log-rank, LASSO, Cox, decision curve analysis, time-dependent Receiver Operating Characteristic, area under the curve (AUC), and likelihood ratio test. P-value <0.05 was considered significant. RESULTS: Follow-up duration was 28.79 ± 22.59 months (median: 25). High-risk patients, determined by the median radiomics score, showed significantly lower survival rates than low-risk patients. Compared with NCCN-IPI, conventional imaging and clinical models, the combined model achieved the highest C-index for both PFS (0.660 internal, 0.802 external) and OS (0.733 internal, 0.781 external) in validation. Net benefit was greater with radiomics than with clinical alone. The combined model exhibited performance with AUCs of 0.680, 0.752, and 0.830 for predicting 1-year, 3-year, and 5-year PFS, and 0.770, 0.789, and 0.863 for OS in internal validation, with PFS AUCs of 0.860 and 0.826 and OS AUCs of 0.859 and 0.748 for 1-year and 3-year survival in external validation. DATA CONCLUSION: Incorporating a multi-sequence MR-based radiomics model into clinical models enhances the assess accuracy for the prognosis of PCNSL. EVIDENCE LEVEL: 4 TECHNICAL EFFICACY: Stage 2.

Automatic segmentation of hepatocellular carcinoma on dynamic contrast-enhanced MRI based on deep learning.

Objective. Precise hepatocellular carcinoma (HCC) detection is crucial for clinical management. While studies focus on computed tomography-based automatic algorithms, there is a rareness of research on automatic detection based on dynamic contrast enhanced (DCE) magnetic resonance imaging. This study is to develop an automatic detection and segmentation deep learning model for HCC using DCE.Approach: DCE images acquired from 2016 to 2021 were retrospectively collected. Then, 382 patients (301 male; 81 female) with 466 lesions pathologically confirmed were included and divided into an 80% training-validation set and a 20% independent test set. For external validation, 51 patients (42 male; 9 female) in another hospital from 2018 to 2021 were included. The U-net architecture was modified to accommodate multi-phasic DCE input. The model was trained with the training-validation set using five-fold cross-validation, and furtherly evaluated with the independent test set using comprehensive metrics for segmentation and detection performance. The proposed automatic segmentation model consisted of five main steps: phase registration, automatic liver region extraction using a pre-trained model, automatic HCC lesion segmentation using the multi-phasic deep learning model, ensemble of five-fold predictions, and post-processing using connected component analysis to enhance the performance to refine predictions and eliminate false positives.Main results. The proposed model achieved a mean dice similarity coefficient (DSC) of 0.81 ± 0.11, a sensitivity of 94.41 ± 15.50%, a precision of 94.19 ± 17.32%, and 0.14 ± 0.48 false positive lesions per patient in the independent test set. The model detected 88% (80/91) HCC lesions in the condition of DSC > 0.5, and the DSC per tumor was 0.80 ± 0.13. In the external set, the model detected 92% (58/62) lesions with 0.12 ± 0.33 false positives per patient, and the DSC per tumor was 0.75 ± 0.10.Significance.This study developed an automatic detection and segmentation deep learning model for HCC using DCE, which yielded promising post-processed results in accurately identifying and delineating HCC lesions.

Joint learning framework of cross-modal synthesis and diagnosis for Alzheimer's disease by mining underlying shared modality information.

Alzheimer's disease (AD) is one of the most common neurodegenerative disorders presenting irreversible progression of cognitive impairment. How to identify AD as early as possible is critical for intervention with potential preventive measures. Among various neuroimaging modalities used to diagnose AD, functional positron emission tomography (PET) has higher sensitivity than structural magnetic resonance imaging (MRI), but it is also costlier and often not available in many hospitals. How to leverage massive unpaired unlabeled PET to improve the diagnosis performance of AD from MRI becomes rather important. To address this challenge, this paper proposes a novel joint learning framework of unsupervised cross-modal synthesis and AD diagnosis by mining underlying shared modality information, improving the AD diagnosis from MRI while synthesizing more discriminative PET images. We mine underlying shared modality information in two aspects: diversifying modality information through the cross-modal synthesis network and locating critical diagnosis-related patterns through the AD diagnosis network. First, to diversify the modality information, we propose a novel unsupervised cross-modal synthesis network, which implements the inter-conversion between 3D PET and MRI in a single model modulated by the AdaIN module. Second, to locate shared critical diagnosis-related patterns, we propose an interpretable diagnosis network based on fully 2D convolutions, which takes either 3D synthesized PET or original MRI as input. Extensive experimental results on the ADNI dataset show that our framework can synthesize more realistic images, outperform the state-of-the-art AD diagnosis methods, and have better generalization on external AIBL and NACC datasets.

A CT-based radiomics integrated model for discriminating pulmonary cryptococcosis granuloma from lung adenocarcinoma-a diagnostic test.

Background: Chest computed tomography (CT) is a critical tool in the diagnosis of pulmonary cryptococcosis as approximately 30% of normal immunity individuals may not exhibit any significant symptoms or laboratory findings. Pulmonary cryptococcosis granuloma and lung adenocarcinoma can appear similar on noncontrast chest CT. This study evaluates the use of an integrated model that was developed based on radiomic features combined with demographic and radiological features to differentiate pulmonary cryptococcosis nodules from lung adenocarcinomas. Methods: Preoperative chest CT images for 215 patients with solid pulmonary nodules with histopathologically confirmed lung adenocarcinoma and cryptococcosis infection were collected from two clinical centers (108 cases in the training set and 107 cases in the test set divided by the different hospitals). Radiomics models were constructed based on nodular lesion volume (LV), 5-mm extended lesion volume (ELV), and perilesion volume (PLV). A demoradiological model was constructed using logistic regression based on demographic information (age, sex) and 12 radiological features (location, number, shape and specific imaging signs). Both models were used to build an integrated model, the performance of which was assessed using the test set. A junior and a senior radiologist evaluated the nodules. Receiver operating characteristic (ROC) curve analysis was conducted, and areas under the curve (AUCs), sensitivity (SEN), and specificity (SPE) of the models were calculated and compared. Results: Among the radiomics models, AUCs of the LV, ELV, and PLV were 0.558, 0.757, and 0.470, respectively. Age, lesion number, and lobular sign were identified as independent discriminative features providing an AUC of 0.77 in the demoradiological model (SEN 0.815, SPE 0.642). The integrated model achieved the highest AUC of 0.801 (SEN 0.759, SPE 0.755), which was significantly higher than that obtained by a junior radiologist (AUC =0.689, P=0.024) but showed no significant difference from that of the senior radiologist (AUC =0.784, P=0.388). Conclusions: An integrated model with radiomics and demoradiological features improves discrimination of cryptococcosis granulomas from solid adenocarcinomas on noncontrast CT. This model may be an effective strategy for machine complementation to discrimination by radiologists, and whole-lung automated recognition methods might dominate in the future.

Modular Level Alterations Of Structural-Functional Connectivity Coupling in Mild Cognitive Impairment Patients and Interactions with Age Effect.

BACKGROUND: Structural-functional connectivity (SC- FC) coupling is related to various cognitive functions and more sensitive for the detection of subtle brain alterations. OBJECTIVE: To investigate whether decoupling of SC-FC was detected in mild cognitive impairment (MCI) patients on a modular level, the interaction effect of aging and disease, and its relationship with network efficiency. METHODS: 73 patients with MCI and 65 healthy controls were enrolled who underwent diffusion tensor imaging and resting-state functional MRI to generate structural and functional networks. Five modules were defined based on automated anatomical labeling 90 atlas, including default mode network (DMN), frontoparietal attention network (FPN), sensorimotor network (SMN), subcortical network (SCN), and visual network (VIS). Intra-module and inter-module SC-FC coupling were compared between two groups. The interaction effect of aging and group on modular SC-FC coupling was further analyzed by two-way ANCOVA. The correlation between the coupling and network efficiency was finally calculated. RESULTS: In MCI patients, aberrant intra-module coupling was noted in SMN, and altered inter-module coupling was found in the other four modules. Intra-module coupling exhibited significant age-by-group effects in DMN and SMN, and inter-module coupling showed significant age-by-group effects in DMN and FPN. In MCI patients, both positive or negative correlations between coupling and network efficiency were found in DMN, FPN, SCN, and VIS. CONCLUSION: SC-FC coupling could reflect the association of SC and FC, especially in modular levels. In MCI, SC-FC coupling could be affected by the interaction effect of aging and disease, which may shed light on advancing the pathophysiological mechanisms of MCI.

Iron quantitative analysis of motor combined with bulbar region in M1 cortex may improve diagnosis performance in ALS.

OBJECTIVES: To explore whether the combined analysis of motor and bulbar region of M1 on susceptibility-weighted imaging (SWI) can be a valid biomarker for amyotrophic lateral sclerosis (ALS). METHODS: Thirty-two non-demented ALS patients and 35 age- and gender-matched healthy controls (HC) were retrospectively recruited. SWI and 3D-T1-MPRAGE images were obtained from all individuals using a 3.0-T MRI scan. The bilateral posterior band of M1 was manually delineated by three neuroradiologists on phase images and subdivided into the motor and bulbar regions. We compared the phase values in two groups and performed a stratification analysis (ALSFRS-R score, duration, disease progression rate, and onset). Receiver operating characteristic (ROC) curves were also constructed. RESULTS: ALS group showed significantly increased phase values in M1 and the two subregions than the HC group, on the all and elderly level (p < 0.001, respectively). On all-age level comparison, negative correlations were found between phase values of M1 and clinical score and duration (p < 0.05, respectively). Similar associations were found in the motor region (p < 0.05, respectively). On both the total (p < 0.01) and elderly (p < 0.05) levels, there were positive relationships between disease progression rate and M1 phase values. In comparing ROC curves, the entire M1 showed the best diagnostic performance. CONCLUSIONS: Combining motor and bulbar analyses as an integral M1 region on SWI can improve ALS diagnosis performance, especially in the elderly. The phase value could be a valuable biomarker for ALS evaluation. KEY POINTS: • Integrated analysis of the motor and bulbar as an entire M1 region on SWI can improve the diagnosis performance in ALS. • Quantitative analysis of iron deposition by SWI measurement helps the clinical evaluation, especially for the elderly patients. • Phase value, when combined with the disease progression rate, could be a valuable biomarker for ALS.

Brain MRI features of anti-N-methyl-D-aspartate (anti-NMDA) receptor encephalitis secondary to central nervous system infection in adult patients.

BACKGROUND: Anti-N-methyl-D-aspartate receptor (anti-NMDAR) encephalitis secondary to central nervous system (CNS) infection is a unique subtype of the autoimmune-mediated disease, of which the imaging features are unclear. PURPOSE: To compare the brain magnetic resonance imaging (MRI) features between the anti-NMDAR encephalitis secondary to CNS infection and that without initial infection. MATERIAL AND METHODS: A total of 70 adult patients with anti-NMDAR encephalitis were retrospectively enrolled (24 in the post-infection group, 46 in the non-infection-related group). Their clinical and imaging features (lesion distribution, lesion shape, enhancement pattern, brain atrophy) were reviewed and summarized. Lesion distributions were compared between the two groups on lesion probability maps. RESULTS: The patients with normal brain MRI scans in the post-infection group were less than those in the non-infection related group (29% vs. 63%; P = 0.0113). Among the 24 patients in the post-infection group, visible lesions were shown at the anti-NMDAR encephalitis onset in 17 patients; lesion distribution was more diffuse than the non-infection-related group, showing higher lesion peak probabilities in the bilateral hippocampus, frontal lobe, temporal lobe, insula, and cingulate. The lesions with contrast enhancement were also more common in the post-infection group than the non-infection-related group (7/13 vs. 2/10). Brain atrophy was observed in eight patients in the post-infection group and three in the non-infection-related group. CONCLUSION: Anti-NMDAR encephalitis secondary to CNS infection has its imaging features-extensive lesion distribution, leptomeningeal enhancement, early atrophy, and necrosis-that could deepen the understanding of the pathophysiology and manifestation of the autoimmune encephalitis besides the classic type.

An MRI-based joint model of radiomics and spatial distribution differentiates autoimmune encephalitis from low-grade diffuse astrocytoma.

Background: The differential diagnosis between autoimmune encephalitis and low-grade diffuse astrocytoma remains challenging. We aim to develop a quantitative model integrating radiomics and spatial distribution features derived from MRI for discriminating these two conditions. Methods: In our study, we included 188 patients with confirmed autoimmune encephalitis (n = 81) and WHO grade II diffuse astrocytoma (n = 107). Patients with autoimmune encephalitis (AE, n = 59) and WHO grade II diffuse astrocytoma (AS, n = 79) were divided into training and test sets, using stratified sampling according to MRI scanners. We further included an independent validation set (22 patients with AE and 28 patients with AS). Hyperintensity fluid-attenuated inversion recovery (FLAIR) lesions were segmented for each subject. Ten radiomics and eight spatial distribution features were selected via the least absolute shrinkage and selection operator (LASSO), and joint models were constructed by logistic regression for disease classification. Model performance was measured in the test set using the area under the receiver operating characteristic (ROC) curve (AUC). The discrimination performance of the joint model was compared with neuroradiologists. Results: The joint model achieved better performance (AUC 0.957/0.908, accuracy 0.914/0.840 for test and independent validation sets, respectively) than the radiomics and spatial distribution models. The joint model achieved lower performance than a senior neuroradiologist (AUC 0.917/0.875) but higher performance than a junior neuroradiologist (AUC 0.692/0.745) in the test and independent validation sets. Conclusion: The joint model of radiomics and spatial distribution from a single FLAIR could effectively classify AE and AS, providing clinical decision support for the differential diagnosis between the two conditions.

A coarse-to-fine cascade deep learning neural network for segmenting cerebral aneurysms in time-of-flight magnetic resonance angiography.

BACKGROUND: Accurate segmentation of unruptured cerebral aneurysms (UCAs) is essential to treatment planning and rupture risk assessment. Currently, three-dimensional time-of-flight magnetic resonance angiography (3D TOF-MRA) has been the most commonly used method for screening aneurysms due to its noninvasiveness. The methods based on deep learning technologies can assist radiologists in achieving accurate and reliable analysis of the size and shape of aneurysms, which may be helpful in rupture risk prediction models. However, the existing methods did not accomplish accurate segmentation of cerebral aneurysms in 3D TOF-MRA. METHODS: This paper proposed a CCDU-Net for segmenting UCAs of 3D TOF-MRA images. The CCDU-Net was a cascade of a convolutional neural network for coarse segmentation and the proposed DU-Net for fine segmentation. Especially, the dual-channel inputs of DU-Net were composed of the vessel image and its contour image which can augment the vascular morphological information. Furthermore, a newly designed weighted loss function was used in the training process of DU-Net to promote the segmentation performance. RESULTS: A total of 270 patients with UCAs were enrolled in this study. The images were divided into the training (N = 174), validation (N = 43), and testing (N = 53) cohorts. The CCDU-Net achieved a dice similarity coefficient (DSC) of 0.616 ± 0.167, Hausdorff distance (HD) of 5.686 ± 7.020 mm, and volumetric similarity (VS) of 0.752 ± 0.226 in the testing cohort. Compared with the existing best method, the DSC and VS increased by 18% and 5%, respectively, while the HD decreased by one-tenth. CONCLUSIONS: We proposed a CCDU-Net for segmenting UCAs in 3D TOF-MRA, and the obtained results show that the proposed method outperformed other existing methods.

Multi-lesion radiomics model for discrimination of relapsing-remitting multiple sclerosis and neuropsychiatric systemic lupus erythematosus.

OBJECTIVES: To develop an MRI-based multi-lesion radiomics model for discrimination of relapsing-remitting multiple sclerosis (RRMS) and its mimicker neuropsychiatric systemic lupus erythematosus (NPSLE). METHODS: A total of 112 patients with RRMS (n = 63) or NPSLE (n = 49) were assigned to training and test sets with a ratio of 3:1. All lesions across the whole brain were manually segmented on T2-weighted fluid-attenuated inversion recovery images. For each single lesion, 371 radiomics features were extracted and trained using machine learning algorithms, producing Radiomics Index for Lesion (RIL) for each lesion and a single-lesion radiomics model. Then, for each subject, single lesions were assigned to one of two disease courts based on their distance to decision threshold, and a Radiomics Index for Subject (RIS) was calculated as the mean RIL value of lesions on the higher-weighted court. Accordingly, a subject-level discrimination model was constructed and compared with performances of two radiologists. RESULTS: The subject-based discrimination model satisfactorily differentiated RRMS and NPSLE in both training (AUC = 0.967, accuracy = 0.892, sensitivity = 0.917, and specificity = 0.872) and test sets (AUC = 0.962, accuracy = 0.931, sensitivity = 1.000, and specificity = 0.875), significantly better than the single-lesion radiomics method (training: p < 0.001; test: p = 0.001) Besides, the discrimination model significantly outperformed the senior radiologist in the training set (training: p = 0.018; test: p = 0.077) and the junior radiologist in both the training and test sets (training: p = 0.008; test: p = 0.023). CONCLUSIONS: The multi-lesion radiomics model could effectively discriminate between RRMS and NPSLE, providing a supplementary tool for accurate differential diagnosis of the two diseases. KEY POINTS: • Radiomic features of brain lesions in RRMS and NPSLE were different. • The multi-lesion radiomics model constructed using a merging strategy was comprehensively superior to the single-lesion-based model for discrimination of RRMS and NPSLE. • The RRMS-NPSLE discrimination model showed a significantly better performance or a trend toward significance than the radiologists.

A novel cascade machine learning pipeline for Alzheimer's disease identification and prediction.

Introduction: Alzheimer's disease (AD) is a progressive and irreversible brain degenerative disorder early. Among all diagnostic strategies, hippocampal atrophy is considered a promising diagnostic method. In order to proactively detect patients with early Alzheimer's disease, we built an Alzheimer's segmentation and classification (AL-SCF) pipeline based on machine learning. Methods: In our study, we collected coronal T1 weighted images that include 187 patients with AD and 230 normal controls (NCs). Our pipeline began with the segmentation of the hippocampus by using a modified U2-net. Subsequently, we extracted 851 radiomics features and selected 37 features most relevant to AD by the Hierarchical clustering method and Least Absolute Shrinkage and Selection Operator (LASSO) algorithm. At last, four classifiers were implemented to distinguish AD from NCs, and the performance of the models was evaluated by accuracy, specificity, sensitivity, and area under the curve. Results: Our proposed pipeline showed excellent discriminative performance of classification with AD vs NC in the training set (AUC=0.97, 95% CI: (0.96-0.98)). The model was also verified in the validation set with Dice=0.93 for segmentation and accuracy=0.95 for classification. Discussion: The AL-SCF pipeline can automate the process from segmentation to classification, which may assist doctors with AD diagnosis and develop individualized medical plans for AD in clinical practice.

Alterations of spontaneous brain activity in systematic lupus erythematosus patients without neuropsychiatric symptoms: A resting-functional MRI study.

PURPOSE: To explore the alterations of spontaneous neuronal activity using amplitude of low-frequency fluctuation (ALFF), fractional amplitude of low-frequency fluctuation (fALFF), and regional homogeneity (ReHo) in non-NPSLE patients and their relationship with the anxiety and depression statuses. METHODS: Twenty-three non-NPSLE patients and 28 healthy controls were enrolled in this study. Resting-state functional magnetic resonance imaging was firstly analyzed by ALFF, fALFF, and ReHo. The relationships between ALFF/fALFF/ReHo values of abnormal regions and anxiety/depression rating scales, including Self-Rating Anxiety (SAS) and Self-Rating Depression (SDS), were also analyzed. RESULTS: Compared with HC, non-NPSLE had decreased ALFF values in the bilateral postcentral gyrus, while increased ALFF values in the bilateral inferior temporal gyrus, left putamen, and bilateral precuneus. Non-NPSLE showed reduced fALFF values in the left lingual gyrus, left middle occipital gyrus, right postcentral gyrus, and left superior parietal gyrus, while increased fALFF values were in the left inferior temporal gyrus, right hippocampus, bilateral precuneus, and bilateral superior frontal gyrus. Reduced ReHo values were in the bilateral postcentral gyrus and higher ReHo values were in the left inferior temporal gyrus, left putamen, and bilateral superior frontal gyrus. In the non-NPSLE group, the mean ALFF values of bilateral precuneus were positively correlated with the SAS rating scales (R = 0.5519, p = 0.0176); either were the mean ALFF values of right inferior temporal gyrus and SAS rating scales (R = 0.5380, p = 0.0213). The mean fALFF values of left inferior temporal gyrus were positively correlated with SAS rating scales (R = 0.5700, p = 0.0135). And the mean ReHo values of left putamen were positively correlated with SDS (R = 0.5477, p = 0.0186). CONCLUSION: Non-NPSLE exhibited abnormal spontaneous neural activity and coherence in several brain regions mainly associated with cognitive and emotional functions. The ALFF values of bilateral PCUN, the right ITG, the fALFF values of left ITG, and the ReHo values of left PUT may be complementary biomarkers for assessing the psychiatric symptoms.

Swallow tail sign on susceptibility map-weighted imaging (SMWI) for disease diagnosing and severity evaluating in parkinsonism.

BACKGROUND: Loss of swallow tail sign (STS) on iron-sensitive magnetic resonance imaging (MRI) has been recognized as an imaging feature in parkinsonism (PS). PURPOSE: To investigate the diagnostic and differential diagnostic value of STS scale on susceptibility map-weighted imaging (SMWI) in PS, including Parkinson's disease (PD), progressive supranuclear palsy syndrome (PSP), and multiple system atrophy (MSA), and to evaluate its correlation with disease severity. MATERIAL AND METHODS: Ninety-nine patients (55 PD, 29 PSP, and 15 MSA) and 47 healthy controls (HC) were prospectively recruited and scanned using quantitative susceptibility mapping (QSM). STS was visually assessed on SMWI derived from QSM. STS scale in the range of 0-4 at participant level was calculated by summing bilateral STS scores (0-2). We used receiver operating characteristic analysis of STS scale for evaluating the diagnostic power of parkinsonism and Spearman's correlation for assessing disease severity. RESULTS: Frequency distribution of STS scale was significantly different in parkinsonism and HC groups, and among PD, PSP, and MSA subgroups. STS scale ≤3 could distinguish parkinsonism from HC with high accuracy (91.78%), PD from HC (91.18%), and MSA from HC (88.71%). STS scale ≤2 could distinguish PSP from HC (96.05%). STS scale = 0 could distinguish PSP from PD (70.24%) and PSP from MSA (72.73%). STS scale was negatively correlated with H-Y stage (P = 0.007, r = -0.359) and duration of disease (P = 0.006, r = -0.367) in PD patients. CONCLUSION: STS scale on SMWI may serve as a useful imaging biomarker for diagnosis of parkinsonism and disease progression evaluation in PD.