| Computer # |
| 4703.
 | 65 | Predicting cognitive performance at age 9 using multimodal MRI neuroimaging at age 7 Isaac Lebogang Khobo1,2, Ernesta Meintjes1,2, Barbara Laughton3, Kaylee van Wyhe3,4, Andre van der Kouwe1,5,6, and Frances Robertson1,2
1Human Biology, University of Cape Town, Cape Town, South Africa, 2Neuroscience Institute UCT, Cape Town, South Africa, 3Paediatrics and Child Health, Stellenbosch University, Family Centre for Research with Ubuntu, Cape Town, South Africa, 4Psychology, Acsent Lab, University of Cape Town, Cape Town, South Africa, 5A.A. Martinos Centre for Biomedical Imaging, Boston, MA, United States, 6Radiology, Harvard Medical School, Boston, MA, United States Keywords: Diagnosis/Prediction, Multimodal Motivation: The relationship between multimodal MRI neuroimaging and future cognitive performance of children from low-socioeconomic status backgrounds remains incompletely understood. Goal(s): We aimed to predict cognitive performance at age 9 using multimodal MRI data of the same children at age 7. Approach: We implemented 10-fold cross validated support vector machines and regression modelling on a combination of structural, diffusion, and spectroscopic MRI to predict continuous scores and categories of cognitive performance. Results: We could predict whether children would fall into a poorer or better scoring category at age 9 with 76% accuracy, 81% specificity, and 72% sensitivity. Impact: We demonstrate the ability to predict overall cognitive
performance at age 9 from neuroimaging 2 years earlier. This could facilitate
identification of at-risk children who may benefit the most from earlier
targeted interventions. |
| 4704.
 | 66 | Segmentation of Multiple Sclerosis lesions using 7T multi-contrast MRI data Anna Petrova1,2, Assunta Dal-Bianco2,3, Eva Niess1, Nik Krajnc2,3, Wolfgang Bogner1, Günther Grabner4, Paul Rommer2,3, and Stanislav Motyka1
1High Filed MR Center, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria, 2Department of Neurology, Medical University of Vienna, Vienna, Austria, 3Comprehensive Center for Clinical Neurosciences & Mental Health, Medical University of Vienna, Vienna, Austria, 4Department of Medical Engineering, Carinthia University of Applied Sciences, Klagenfurt, Austria Keywords: Diagnosis/Prediction, Machine Learning/Artificial Intelligence Motivation: The effective treatment of Multiple sclerosis (MS) requires reliable estimates of lesion load and hence precise lesion detection over time. However, current lesion load estimation is either qualitative or too time-consuming.
Goal(s): Our study automates MS lesion segmentation by training DeepMedic for application to 7T multi-contrast MRI data of MS patients.
Approach: Training with all four contrasts achieved the best results compared to Lesion Segmentation Tool (LST)—a conventional/non-deep-learning SPM-based MS lesions segmentation approach.
Results: Our study highlights potential for automating MS lesion detection/segmentation for 7T multi-contrast MRI data, underscoring the importance of accurate ground truth data and high-quality databases for improved detection accuracy.
Impact: The
results of this research will impact the user-independent
detection/segmentation of multiple sclerosis lesions, making manual
assessment by clinicians obsolete and enable fully automated
monitoring of lesions load as a quantitative radiological marker of
disease progression. |
| 4705.
 | 67 | Prediction of myelin stainings using 7T MRI and deep learning Sutatip Pittayapong1,2, Simon Hametner2, Romana Höftberger2, and Grabner Günther1
1Department of Medical Engineering, Carinthia University of Applied Sciences, Klagenfurt, Austria, 2Division of Neuropathology and Neurochemistry, Department of Neurology, Medical University of Vienna, Vienna, Austria Keywords: Diagnosis/Prediction, Machine Learning/Artificial Intelligence, Brain Histological Prediction Motivation: Histological examination of the brain provides precise analyses of brain tissue, but requires ex vivo tissue samples. In vivo MRI offers an alternative but still has limitations. Goal(s): Predict histologic images from in vivo MR images. Approach: Utilize deep learning generative models to create brain histological images from multi contrast MR images. Results: The appropriate combination of MRI contrasts can generate myelin histology images. Impact: Generating myelin histological images significantly impacts brain tissue property research by providing ex vivo information. This adaptable technique extends its applicability to various tissue properties, providing broader insights into histology and beyond. |
| 4706.
 | 68 | Macromolecules from short echo time 7 Tesla brain 1HMRS as biomarkers of the Alzheimer’s disease continuum Andrea Dell'Orco1,2,3,4, Laura Göschel2,3, Layla Tabea Riemann4,5, Semiha Aydin4, Bernd Ittermann4, Anna Tietze1, Michael Scheel1, and Ariane Fillmer4
1Department of Neuroradiology, Charité – Universitätsmedizin Berlin, Berlin, Germany, 2Department of Neurology, Charité – Universitätsmedizin Berlin, Berlin, Germany, 3NeuroScience Clinical Research Center, Charité – Universitätsmedizin Berlin, Berlin, Germany, 4Physikalisch-Technische Bundesanstalt (PTB), Berlin, Germany, 5Institute for Applied Medical Informatics, University Hospital Hamburg-Eppendorf (UKE), Hamburg, Germany Keywords: Diagnosis/Prediction, Spectroscopy Motivation: To investigate the potential of macromolecule (MM) from 1HMRS as biomarkers for Alzheimer's Disease (AD). Goal(s): Enhance the MRS-only diagnostic prediction for the AD continuum by incorporating MM data. Approach: We predict the diagnosis of 143 individuals ranging from cognitively healthy to AD using only 1HMRS data, employing OPLSDA. We compare the model's performance with/without MM and validate the results with a second ML classifier. We also evaluate variable importance in classification. Results: The inclusion of MM signals improves AD diagnosis prediction when OPLSDA is used. Various MM peaks contribute to the classification. However, the transitional stage of MCI cannot be accurately classified. Impact: When combined with the appropriate
method, MM signals can enhance the diagnosis of AD using MRS as a stand-alone
marker, and important MM peaks belonging to the AD neurochemical fingerprint
were identified. |
| 4707.
 | 69 | An Alzheimer's Disease Progression Score Using Supervised Variational Autoencoders on MRI Anatomic Imaging Data Junhyoun Sung1, Dean Shibata2,3, Kwun Chuen Gary Chan1,3, Lan Shui3,4, and David Haynor2
1Department of Biostatistics, University of Washington, Seattle, WA, United States, 2Department of Radiology, University of Washington, Seattle, WA, United States, 3National Alzheimer’s Coordinating Center, Seattle, WA, United States, 4Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX, United States Keywords: Diagnosis/Prediction, Alzheimer's Disease Motivation: Alzheimer's disease affects millions, but understanding its progression remains challenging. This study seeks to assess the severity of Alzheimer's from imaging data alone. Goal(s): To create a score that reflects how far Alzheimer's has progressed in a patient. Approach: Using brain scans and simple patient demographic information, we developed an imaging-based model that predicts the severity of Alzheimer's. Results: Our model successfully distinguishes between different stages of Alzheimer's, offering a reliable disease progression score. Impact: This work could lead to
earlier detection and better tracking of Alzheimer's, informing treatment
decisions and aiding in the objective development and evaluation of new
therapies. |
| 4708.
 | 70 | Estimating brain age and identifying important neural connections in mouse models of genetic risk for Alzheimer’s disease Hae Sol Moon1, Ali Mahzarnia2, Jacques Stout3, Robert J Anderson2, Zay Yar Han2, and Alexandra Badea1,2,3,4
1Biomedical Engineering, Duke University, Durham, NC, United States, 2Radiology, Duke University School of Medicine, Durham, NC, United States, 3Brain Imaging and Analysis Center, Duke University School of Medicine, Durham, NC, United States, 4Neurology, Duke University School of Medicine, Durham, NC, United States Keywords: Diagnosis/Prediction, Alzheimer's Disease, Diffusion MRI Motivation: Understanding Alzheimer’s disease (AD) requires decoding the complex interplay of risk factors, particularly how age-related structural connectivity changes affect AD onset and progression. Goal(s): Using a state-of-the-art deep learning method, we aim to identify key brain connections, predict age and assess AD risk factors with structural brain connectomes and behavioral data from mouse models with humanized APOE genotypes. Approach: Our Feature Attention Graph Neural Network (FAGNN) integrates multivariate data types, focusing on aging-related brain connections with a quadrant attention module. Results: FAGNN surpassed other models in age prediction and identified critical neural pathways, like striatum-cingulum connection, offering insights into age-related brain connectivity changes. Impact: We used AI and FAGNN to advance Alzheimer’s disease research, predicting risk factors such as age and identifying crucial neural connections pertinent to the risk factors, potentially paving the way for early detection and targeted interventions in aging-related cognitive decline. |
| 4709.
 | 71 | Deep learning-based estimation of future brain atrophy using baseline MRI and PET Linh N. N. Le1, Evan Fletcher2, Jinyi Qi1, and Audrey P. Fan1,2
1Biomedical Engineering, University of California, Davis, Davis, CA, United States, 2Neurology, University of California, Davis, Davis, CA, United States Keywords: Diagnosis/Prediction, Alzheimer's Disease, Deep learning Motivation: Prediction of brain atrophy, a key feature in Alzheimer’s Disease (AD), is critical to observe disease progression before the onset of irreversible atrophy. Goal(s): We aim to predict future cortical atrophy rates in the elderly population from baseline PET and MRI scans. Approach: We predict image-derived cortical atrophy rate as an anatomical biomarker of neurodegeneration using image-generation deep learning networks based on T1-weighted structural MRI and PET as inputs. Results: Both T1- and PET-based models can predict longitudinal atrophy maps from baseline, with greater average atrophy in AD and mild cognitive impairment compared to cognitively normal, consistent with the Tensor-Based Morphometry method. Impact: Predicting future brain atrophy from baseline imaging can show disease progression before the onset of irreversible atrophy. Early detection of cognitive impairment and Alzheimer’s Disease progression would support planning for patient care and monitoring new lifestyle interventions and pharmacological therapies. |
| 4710.
 | 72 | Automated Deep Learning-Based Magnetic Resonance Parkinsonism Index 2.0 in Early Parkinson’s Disease: A Longitudinal Study Septian Hartono1,2,3, Punith B Venkategowda4,5, Madappa S4, Ricardo Corredor Jerez6,7,8, Bénédicte Maréchal6,7,8, Tommaso Di Noto6,7,8, Samuel Yong Ern Ng1, Nicole Shuang Yu Chia1, Yiu Cho Chung9, Julian Gan9, Louis Chew Seng Tan1,2, Eng King Tan1,2, and Ling Ling Chan2,3
1National Neuroscience Institute, Singapore, Singapore, 2Duke-NUS Medical School, Singapore, Singapore, 3Singapore General Hospital, Singapore, Singapore, 4Siemens Healthineers India, Bangalore, India, 5International Institute of Information Technology, Bangalore, India, 6Siemens Healthineers International AG, Lausanne, Switzerland, 7École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 8Lausanne University Hospital, Lausanne, Switzerland, 9Siemens Healthineers Singapore, Singapore, Singapore Keywords: Analysis/Processing, Machine Learning/Artificial Intelligence Motivation: There is a pressing need for accurate and reliable methods to track disease progression in patients presenting with early Parkinsonism. Goal(s): We aim to assess the utility of a fully automated deep learning-based method developed to estimate both MRPI 1.0 and 2.0 measures in a large, longitudinal, three time-point case-control cohort of patients presenting with early Parkinsonism. Approach: MRPI 1.0 and 2.0 measures were computed from 3D T1-weighted images using the Quantitative Brain Assessment Toolkit (QBAT, v2.1.0) research application. Results: MRPI 2.0 showed improved group differentiation and disease classification when compared to MRPI 1.0 in patients presenting with early Parkinsonism. Impact: Automated, deep learning-based MRPI 2.0 assessment may be used as quick tool to facilitate radiological screening, complementary to other quantitative MRI techniques such as quantitative susceptibility mapping and diffusion MRI, to track progression in Parkinson’s disease. |
| 4711.
 | 73 | CNN-based Automated Pipeline for Accurate Computation of Magnetic Resonance Parkinsonism’s Index Measurements Punith B Venkategowda1,2, Tommaso Di Noto3,4,5, Ricardo Corredor-Jerez3,4,5, Tobias Bodenmann3, Madappa Shadakshari Swamy1, Bhairav Mehta1, Alessandra Griffa6, Sandrine Nadeau6, Gilles Allali6, Ling Ling Chan7,8, Vincent Dunet5, Jitender Saini9, Max Scheffler10, Neelam Sinha2, and Bénédicte Maréchal3,4,5
1Siemens Healthineers India, Bengaluru, India, Bengaluru, India, 2International Institute of Information Technology Bangalore, Bengaluru, India, 3Advanced Clinical Imaging Technology, Siemens Healthineers International AG, Lausanne, Switzerland, 4LTS5, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 5Department of Diagnostic and Interventional Radiology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland, 6Leenaards Memory Centre, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland, 7Singapore General Hospital, Singapore, Singapore, 8Duke-NUS Medical School, Singapore, Singapore, 9National Institute of Mental Health and Neurosciences, Bengaluru, India, 10Division of Radiology, Geneva University Hospitals, Geneva, Switzerland Keywords: Diagnosis/Prediction, Neurodegeneration, MRPI, Progressive supranuclear palsy, Parkinsonism Index, CAD Motivation: Magnetic resonance parkinsonism index (MRPI) has shown promising results in differentiating progressive supranuclear palsy from idiopathic Parkinson’s disease and the Parkinson variant of multiple system atrophy (MSA-P). Goal(s): In this work, we propose a fully automated pipeline to calculate MRPI using a convolutional neural network (CNN). This can be a time-saving tool in making diagnoses in clinically ambiguous cases. Approach: Our method utilizes registration and deep learning-based segmentation techniques to extract relevant measurements from T1 weighted MRI images (T1w). Results: Experimental results demonstrated the robustness of our approach and its generalizability across different clinical settings. Impact: Automating the measurement of MRPI components with a deep learning based algorithm can help providing objective and reproducible measures. It may be beneficial for differential diagnosis of patients with Parkinsonian syndromes with significant savings in reporting time. |
| 4712.
 | 74 | An Automatic Striatum Segmentation Model to Estimate MR Maps for Dopamine Transporter SPECT using Deep Learning Haiyan Wang1,2, Han Jiang2,3, Gefei Chen2, Yu Du2,4, Zhonglin Lu2,4, Hairong Zheng1, Dong Liang1, Greta S. P. Mok2,4, and Zhanli Hu1
1Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, 2Biomedical Imaging Laboratory (BIG), Department of Electrical and Computer Engineering, Faculty of Science and Technology, University of Macau, Macau SAR, China, 3PET-CT Center, Fujian Medical University Union Hospital, Fuzhou, China, 4Center for Cognitive and Brain Sciences, Institute of Collaborative Innovation, University of Macau, Macau SAR, China Keywords: AI/ML Software, Parkinson's Disease, Cross-modality, Deep learning, SPECT, Striatum, Segmentation Motivation: Striatum segmentation on SPECT is necessary to quantify uptake for Parkinson's disease (PD), but is challenging due to the inferior resolution. MRI is the preferred reference for segmentation due to its excellent soft tissue contrast. Goal(s): This work proposes cross-modality automatic striatum segmentation, estimating MR striatal maps from clinical SPECT using deep learning (DL). Approach: nnU-Net-based method are implemented and SPECT images are paired with MR-based striatal maps as supervised learning (training:validation:testing = 136:24:40) Results: The proposed method can segment 4 MR-like individual compartments on clinical SPECT, which is also superior to several traditional and DL methods, both in physical and clinical metrics. Impact: The proposed DL-based cross-modality striatum segmentation method
is feasible for clinical DaT SPECT in PD, and 4 MR-like individual compartments
can be obtained to quantify striatal uptake, which is beneficial to the
accurate diagnosis and clinical management of PD. |
| 4713.
 | 75 | Prediction of Lifetime Acute Ischemic Stroke Risk using Multimodal Models and SHAP-based Interpretability Methods Wenyue Mao1, Yuxiang Dai1, Zhang Shi2, Rencheng Zheng1, Yinghua Chu3, Chengyan Wang4, and He Wang1,4
1Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, China, 2Department of Radiology, Zhongshan Hospital, Fudan University, Shanghai, China, 3Simens Healthineers Ltd., Shanghai, China, 4Human Phenome Institute, Fudan University, Shanghai, China Keywords: Diagnosis/Prediction, Stroke, Stroke occurence prediction; Deep learning Motivation:
Accurately predicting the lifetime risk of acute ischemic stroke (AIS) remains a significant challenge, and there is a notable scarcity of multimodal models that effectively integrate medical imaging with clinical factors. Goal(s):
To propose an effective multi-modality deep learning model based on both MR images and clinical factors for improved prediction of AIS occurrence. Approach:
The model leveraged a clinical-factor-agnostic module to extract clinical features from clinical factors and employed Shapley methods to scrutinize the significance of features. Results:
The proposed method achieved higher performance than conventional models for the prediction of lifetime AIS occurrence. Impact: Our model's predictive outcomes could pinpoint
individuals at high risk for AIS, allowing clinicians to advise them on
self-health vigilance. |
| 4714.
 | 76 | A Large-Scale Analysis of the Impact of Physical Exercise, BMI, and Lifestyle on Brain Age Predicted from T1-weighted MRI Scans Soojin Lee1,2, Saurabh Garg1,2, Saqib Basar1,2, Thanh-Duc Nguyen1,2, Nasrin Akbari1,2, Madhurima Datta1,2, Arun Rajendran1,2, Yosef Chodakiewitz2, Kellyann Niotis3,4, Rajpaul Attariwala1,2, and Sam Hashemi1,2
1VoxelWise Imaging Technology Inc, Vancouver, BC, Canada, 2Prenuvo Inc, Vancouver, BC, Canada, 3Early Medical, Austin, TX, United States, 4The Institute of Neurodegenerative Diseases of Florida, Boca Raton, FL, United States Keywords: Diagnosis/Prediction, Brain Motivation: To study the effects of exercise, BMI and lifestyle factors on brain age. Goal(s): Developing a model for predicting brain age based on T1-weighted MRI scans. Approach: T1-weighted MRI scans of 8,770 individuals were examined. Normative brain age curves were generated with over 50,000 volumetric brain MRI scans. Results: Hypertension, type 2 diabetes, and smoking were associated with increased brain age, while exercise significantly decreased it. Pronounced effects of exercise were found in the overweight group, suggesting an increased benefit. The findings emphasize the importance of exercise in preserving brain volumes likely providing neuroprotective effects. Impact: Leveraging a brain age estimation model, we revealed protective effects of exercise on the aging brain, particularly pronounced in overweight individuals. This highlights the potential of brain age as a biomarker for monitoring and developing strategies to enhance brain health. |
| 4715.
 | 77 | Prediction patients with alcohol dependence via graph classification on Brain Network Derived from Functional Magnetic Resonance Imaging Shin-Eui Park1, Yeong-Jae Jeon1,2, and Hyeon-Man Baek1,2
1Lee Gil Ya Cancer & Diabetes Institute, Gachon University, Incheon, Korea, Republic of, 2Department of Health Sciences and Technology, GAIHST, Gachon University, Incheon, Korea, Republic of Keywords: Diagnosis/Prediction, fMRI (resting state), Alcohol dependence, Graph classification, graph embedding technique Motivation: The study is motivated by the need for innovative approaches to alcoholism classification, leveraging neuro-functional network analysis from fMRI data to improve diagnostic accuracy and gain insights into alcoholism's complex nature. Goal(s): The primary goal is to achieve accurate alcoholism classification using functional connectivity patterns and machine learning. Approach: The study employed fMRI data from 15 healthy controls and 15 patients with alcohol dependence, utilizing advanced graph analysis techniques and machine learning algorithms. Results: The approach demonstrated a 73% classification accuracy, highlighting the potential of functional connectivity patterns as diagnostic markers and the value of machine learning in quantifying network differences. Impact: This research contributes to more precise
alcoholism diagnosis and offers opportunities for biomarker discovery. It may
facilitate earlier intervention and more effective treatments, benefiting both
clinicians and patients. The impact includes advancing addiction research and
improving patient care. |
| 4716.
 | 78 | Diagnosis of Early Mild Cognitive Impairment in Type 2 Diabetes Mellitus by Deep Learning of Multimodal Neuroimages and Metadata Kangfu Han1,2, Xiaomei Yue2,3, Shijun Qiu3, Feng Yang1, and Gang Li2
1Southern Medical University, Guangzhou, China, 2University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, 3Guangzhou University of Chinese Medicine, Guangzhou, China Keywords: Diagnosis/Prediction, Brain Motivation: Identification of early cognitive impairment in type 2 diabetes mellitus (T2DM) patients is of paramount importance for mitigating cognitive decline of patients and enhancing their quality of life. Goal(s): Our objective was to develop a robust deep learning model for diagnosing early cognitive impairment in T2DM using multi-modal neuroimages. Approach: We developed a multi-modal neural network, which incorporated informative clinical metadata (i.e., MoCA, BMI and HbA1c) to design metadata-induced contrastive Laplacian regularization. Results: The proposed approach demonstrated significant improvement in accuracy in the identification of T2DM with/without mild cognitive impairment in a dataset with 311 subjects. Impact: Superior diagnostic performance of the proposed
method for early cognitive impairment in T2DM demonstrates its ability in
understanding of T2DM cognitive impairment associated brain alterations and its
potential applications on other brain disorders. |
| 4717.
 | 79 | Graph Neural Networks Elucidating Mesial Temporal Lobe Epilepsy Through Simultaneously Acquired PET and DTI Data – A Pilot Study Tianyun Zhao1,2, Jia Ying1,2, Siyu Yuan3, Hui Huang3, Miao Zhang4, Jie Luo3, and Chuan Huang1,5
1Radiology and Imaging Science, Emory University School of Medicine, Atlanta, GA, United States, 2Biomedical Engineering, Stony Brook University, Stony Brook, NY, United States, 3School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, 4Ruijin Hospital, Shanghai, China, 5Biomedical Engineering, Georgia Institute of Technology, Atalnta, GA, United States Keywords: Analysis/Processing, Epilepsy, AI, MTLE Motivation: There is an urgent need to improve diagnostic accuracy and surgical outcomes for Mesial Temporal Lobe Epilepsy (MTLE) patients, particularly those with drug-resistant forms and unclear epileptogenic zones
Goal(s): To explore the relation between structural connectivity and FDG PET uptake by using Graph Neural Network.
Approach: Graphs were constructed based on diffusion images of the patients. A graph network was trained to predict FDG PET uptake in selected regions. Results: The graph network was able to predict FDG uptake in several regions such as thalamus, middle temporal, and entorhinal cortex. Whereas the network failed to predict uptake in some other regions.
Impact: The study advances understanding of the underlying mechanisms in MTLE by illuminating the relationship between white matter structural connectivity and regional metabolic activity, which could lead to enhanced diagnostic approaches and targeted therapies. |
| 4718.
 | 80 | Radiomics for Deep Brain Stimulation outcome prediction using Quantitative Susceptibility Mapping (RadDBS-QSM) Alexandra Grace Roberts1, Jinwei Zhang2, Heejong Kim3, Dominick Romano4, Sema Akkus5, Mert Sabuncu1,3, Jianqi Li6, Brian Harris Kopell5, Pascal Spincemaille3, and Yi Wang3,4
1Electrical and Computer Engineering, Cornell University, New York, NY, United States, 2Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, United States, 3Radiology, Weill Cornell Medicine, New York, NY, United States, 4Biomedical Engineering, Cornell University, New York, NY, United States, 5Neurosurgery, Mount Sinai Hospital, New York, NY, United States, 6Changhai Hospital, Shanghai, China Keywords: Diagnosis/Prediction, Radiomics Motivation: To improve outcome prediction for deep brain stimulation (DBS) surgery using radiomic features on quantitative susceptibility maps (QSMs). Goal(s): To address the inconsistent levodopa challenge test (LCT) prediction for DBS outcomes by describing the target variable, motor symptom improvement, as a weighted sum of QSM radiomic features. Approach: A least absolute shrinkage and selection operator (LASSO) model is implemented, trained, and tested on patient data and known DBS outcomes. Results: Model predictions outperform the conventional LCT prediction and estimate DBS improvement from preoperative motor symptom scores and radiomic features on QSM. Impact: The levodopa
challenge test estimates patient response to deep brain stimulation surgery, presenting
undesirable side effects and inconsistent outcomes. Radiomic prediction of deep
brain surgery outcomes using quantitative susceptibility maps aims to provide a
numerical measure of symptom improvement. |