1
|
Weng T, Zheng Y, Xie Y, Qin W, Guo L. Diagnosing schizophrenia using deep learning: Novel interpretation approaches and multi-site validation. Brain Res 2024; 1833:148876. [PMID: 38513996 DOI: 10.1016/j.brainres.2024.148876] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/04/2023] [Revised: 02/28/2024] [Accepted: 03/18/2024] [Indexed: 03/23/2024]
Abstract
Schizophrenia is a profound and enduring mental disorder that imposes significant negative impacts on individuals, their families, and society at large. The development of more accurate and objective diagnostic tools for schizophrenia can be expedited through the employment of deep learning (DL), that excels at deciphering complex hierarchical non-linear patterns. However, the limited interpretability of deep learning has eroded confidence in the model and restricted its clinical utility. At the same time, if the data source is only derived from a single center, the model's generalizability is difficult to test. To enhance the model's reliability and applicability, leave-one-center-out validation with a large and diverse sample from multiple centers is crucial. In this study, we utilized Nine different global centers to train and test the 3D Resnet model's generalizability, resulting in an 82% classification performance (area under the curve) on all datasets sourced from different countries, employing a leave-one-center-out-validation approach. Per our approximation of the feature significance of each region on the atlas, we identified marked differences in the thalamus, pallidum, and inferior frontal gyrus between individuals with schizophrenia and healthy controls, lending credence to prior research findings. At the same time, in order to translate the model's output into clinically applicable insights, the SHapley Additive exPlanations (SHAP) permutation explainer method with an anatomical atlas have been refined, thereby offering precise neuroanatomical and functional interpretations of different brain regions.
Collapse
Affiliation(s)
- Tingting Weng
- School of Medical Imaging, Tianjin Medical University, Tianjin 300203, China
| | - Yuemei Zheng
- Department of Medical Imaging, Affiliated Hospital of Jining Medical University, Shandong 100038, China
| | - Yingying Xie
- Department of Radiology and Tianjin Key Laboratory of Functional Imaging, Tianjin Medical University General Hospital, Tianjin 300052, China
| | - Wen Qin
- Department of Radiology and Tianjin Key Laboratory of Functional Imaging, Tianjin Medical University General Hospital, Tianjin 300052, China
| | - Li Guo
- School of Medical Imaging, Tianjin Medical University, Tianjin 300203, China.
| |
Collapse
|
2
|
He Y, Wang Z, Zuo M, Zhang S, Li W, Chen S, Yuan Y, Yang Y, Liu Y. The impact of neurocognitive and psychiatric disorders on the risk of idiopathic normal pressure hydrocephalus: A bidirectional Mendelian randomization study. Brain Behav 2024; 14:e3532. [PMID: 38779749 PMCID: PMC11112403 DOI: 10.1002/brb3.3532] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/22/2023] [Revised: 04/17/2024] [Accepted: 04/17/2024] [Indexed: 05/25/2024] Open
Abstract
BACKGROUND Neurocognitive and psychiatric disorders have been proved that they can comorbid more often with idiopathic normal pressure hydrocephalus (iNPH) than general population. However, the potential causal association between these disorders and iNPH has not been assessed. Thus, our study aims to investigate the causal relationship between them based on a bidirectional Mendelian randomization (MR) analysis. METHODS Random effects of the inverse variance weighted (IVW) method were conducted to obtain the causal association among the neurocognitive disorders, psychiatric disorders, and iNPH. Genome-wide association studies (GWAS) of 12 neurocognitive and psychiatric disorders were downloaded via the OpenGWAS database, GWAS Catalog, and Psychiatric Genomics Consortium, whereas GWAS data of iNPH were obtained from the FinnGen consortium round 9 release, with 767 cases and 375,610 controls of European ancestry. We also conducted the sensitivity analysis in these significant causal inferences using weighted median model, Cochrane's Q test, MR-Egger regression, MR Pleiotropy Residual Sum and Outlier detect and the leave-one-out analysis. RESULTS For most of the neurocognitive and psychiatric disorders, no causal association was established between them and iNPH. We have found that iNPH (odds ratio [OR] = 1.030, 95% confidence interval [CI]: 1.011-1.048, p = .001) is associated with increased risk for schizophrenia, which failed in validation of sensitivity analysis. Notably, genetically predicted Parkinson's disease (PD) is associated with increased risk of iNPH (OR = 1.256, 95% CI: 1.045-1.511, p = .015). CONCLUSION Our study has revealed the potential causal effect in which PD associated with an increased risk of iNPH. Further study is warranted to investigate the association between PD and iNPH and the potential underlying mechanism.
Collapse
Affiliation(s)
- Yuze He
- Department of NeurosurgeryWest China HospitalSichuan UniversityChengduChina
| | - Zhihao Wang
- Department of NeurosurgeryWest China HospitalSichuan UniversityChengduChina
| | - Mingrong Zuo
- Department of Pediatric NeurosurgeryWest China Women's and Children's Hospital, Sichuan University West China Second University HospitalChengduChina
| | - Shuxin Zhang
- Department of Critical Care MedicineWest China HospitalSichuan UniversityChengduChina
| | - Wenhao Li
- Department of NeurosurgeryWest China HospitalSichuan UniversityChengduChina
| | - Siliang Chen
- Department of NeurosurgeryWest China HospitalSichuan UniversityChengduChina
| | - Yunbo Yuan
- Department of NeurosurgeryWest China HospitalSichuan UniversityChengduChina
| | - Yuan Yang
- Department of NeurosurgeryWest China HospitalSichuan UniversityChengduChina
| | - Yanhui Liu
- Department of NeurosurgeryWest China HospitalSichuan UniversityChengduChina
| |
Collapse
|
3
|
Boyd ED, Kaur J, Ding G, Chopp M, Jiang Q. Clinical magnetic resonance imaging evaluation of glymphatic function. NMR IN BIOMEDICINE 2024:e5132. [PMID: 38465514 DOI: 10.1002/nbm.5132] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/07/2023] [Revised: 01/31/2024] [Accepted: 02/06/2024] [Indexed: 03/12/2024]
Abstract
The glymphatic system is a system of specialized perivascular spaces in the brain that facilitates removal of toxic waste solutes from the brain. Evaluation of glymphatic system function by means of magnetic resonance imaging (MRI) has thus far been largely focused on rodents because of the limitations of intrathecal delivery of gadolinium-based contrast agents to humans. This review discusses MRI methods that can be employed clinically for glymphatic-related measurements intended for early diagnosis, prevention, and the treatment of various neurological conditions. Although glymphatic system-based MRI research is in its early stages, recent studies have identified promising noninvasive MRI markers associated with glymphatic system alterations in neurological diseases. However, further optimization in data acquisition, validation, and modeling are needed to investigate the glymphatic system within the clinical setting.
Collapse
Affiliation(s)
- Edward D Boyd
- Department of Neurology, Henry Ford Health System, Detroit, Michigan, USA
- Department of Radiology, Michigan State University, East Lansing, Michigan, USA
| | - Jasleen Kaur
- Department of Neurology, Henry Ford Health System, Detroit, Michigan, USA
- Department of Physics, Oakland University, Rochester, Michigan, USA
| | - Guangliang Ding
- Department of Neurology, Henry Ford Health System, Detroit, Michigan, USA
- Department of Radiology, Michigan State University, East Lansing, Michigan, USA
| | - Michael Chopp
- Department of Neurology, Henry Ford Health System, Detroit, Michigan, USA
- Department of Physics, Oakland University, Rochester, Michigan, USA
| | - Quan Jiang
- Department of Neurology, Henry Ford Health System, Detroit, Michigan, USA
- Department of Radiology, Michigan State University, East Lansing, Michigan, USA
- Department of Physics, Oakland University, Rochester, Michigan, USA
| |
Collapse
|
4
|
Mehta NH, Suss RA, Dyke JP, Theise ND, Chiang GC, Strauss S, Saint-Louis L, Li Y, Pahlajani S, Babaria V, Glodzik L, Carare RO, de Leon MJ. Quantifying cerebrospinal fluid dynamics: A review of human neuroimaging contributions to CSF physiology and neurodegenerative disease. Neurobiol Dis 2022; 170:105776. [PMID: 35643187 PMCID: PMC9987579 DOI: 10.1016/j.nbd.2022.105776] [Citation(s) in RCA: 15] [Impact Index Per Article: 7.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/07/2022] [Accepted: 05/21/2022] [Indexed: 01/13/2023] Open
Abstract
Cerebrospinal fluid (CSF), predominantly produced in the ventricles and circulating throughout the brain and spinal cord, is a key protective mechanism of the central nervous system (CNS). Physical cushioning, nutrient delivery, metabolic waste, including protein clearance, are key functions of the CSF in humans. CSF volume and flow dynamics regulate intracranial pressure and are fundamental to diagnosing disorders including normal pressure hydrocephalus, intracranial hypotension, CSF leaks, and possibly Alzheimer's disease (AD). The ability of CSF to clear normal and pathological proteins, such as amyloid-beta (Aβ), tau, alpha synuclein and others, implicates it production, circulation, and composition, in many neuropathologies. Several neuroimaging modalities have been developed to probe CSF fluid dynamics and better relate CSF volume and flow to anatomy and clinical conditions. Approaches include 2-photon microscopic techniques, MRI (tracer-based, gadolinium contrast, endogenous phase-contrast), and dynamic positron emission tomography (PET) using existing approved radiotracers. Here, we discuss CSF flow neuroimaging, from animal models to recent clinical-research advances, summarizing current endeavors to quantify and map CSF flow with implications towards pathophysiology, new biomarkers, and treatments of neurological diseases.
Collapse
Affiliation(s)
- Neel H Mehta
- Department of Biology, Cornell University, Ithaca, NY, USA
| | - Richard A Suss
- Division of Neuroradiology, Department of Radiology, University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - Jonathan P Dyke
- Citigroup Biomedical Imaging Center, Weill Cornell Medicine, New York, NY, USA
| | - Neil D Theise
- Department of Pathology, NYU Grossman School of Medicine, New York, NY, USA
| | - Gloria C Chiang
- Division of Neuroradiology, Department of Radiology, Weill Cornell Medicine, New York, NY, USA
| | - Sara Strauss
- Division of Neuroradiology, Department of Radiology, Weill Cornell Medicine, New York, NY, USA
| | | | - Yi Li
- Brain Health Imaging Institute, Department of Radiology, Weill Cornell Medicine, New York, NY, USA
| | - Silky Pahlajani
- Brain Health Imaging Institute, Department of Radiology, Weill Cornell Medicine, New York, NY, USA
| | - Vivek Babaria
- Orange County Spine and Sports, Interventional Physiatry, Newport Beach, CA, USA
| | - Lidia Glodzik
- Brain Health Imaging Institute, Department of Radiology, Weill Cornell Medicine, New York, NY, USA
| | - Roxana O Carare
- Department of Medicine, University of Southampton, Southampton, UK
| | - Mony J de Leon
- Brain Health Imaging Institute, Department of Radiology, Weill Cornell Medicine, New York, NY, USA.
| |
Collapse
|