| Literature DB >> 36238504 |
So Yeong Jeong, Chong Hyun Suh, Ho Young Park, Hwon Heo, Woo Hyun Shim, Sang Joon Kim.
Abstract
The incidence of neurodegenerative diseases in the older population has increased in recent years. A considerable number of studies have been performed to characterize these diseases. Imaging analysis is an important biomarker for the diagnosis of neurodegenerative disease. Objective and reliable assessment and precise detection are important for the early diagnosis of neurodegenerative diseases. Artificial intelligence (AI) using brain MRI applied to the study of neurodegenerative diseases could promote early diagnosis and optimal decisions for treatment plans. MRI-based AI software have been developed and studied worldwide. Representatively, there are MRI-based volumetry and segmentation software. In this review, we present the development process of brain volumetry analysis software in neurodegenerative diseases, currently used and developed AI software for neurodegenerative disease in the Republic of Korea, probable uses of AI in the future, and AI software limitations. CopyrightsEntities:
Year: 2022 PMID: 36238504 PMCID: PMC9514516 DOI: 10.3348/jksr.2022.0048
Source DB: PubMed Journal: Taehan Yongsang Uihakhoe Chi ISSN: 1738-2637
Research on the Correlation and Diagnostic Performance of NeuroQuant®
| Author | Year | Journal | Population | Control | Correlation | Diagnostic Performance | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Comparison | Coefficient | Method | Sensitivity (%) | Specificity (%) | AUC | |||||
| Brewer et al. ( | 2009 | AJNR Am J Neuroradiol | 20 mild probable AD | 20 healthy | Manual segmentation | ICC, 0.61–0.99 | ||||
| Ochs et al. ( | 2015 | J Neuroimaging | 20 AD | 20 healthy | FreeSurfer | ICC, 0.62–0.99 | ||||
| Tanpitukpongse et al. ( | 2017 | AJNR Am J Neuroradiol | 85 MCI to AD | 107 MCI nonconverter | Neuroreader | r, 0.60–0.99 | Hippocampus | 0.69 | ||
| Min et al. ( | 2017 | AJR Am J Roentgenol | 30 mild-to-moderate AD | 25 healthy | MTL index A | 63.3 | 100 | 0.83 | ||
| Persson et al. ( | 2018 | Acta Neurol Scand | 31 AD | 14 SCI, | MTA score | r, -0.77 | Hippocampus | 71 | 92 | 0.88 |
| Lee et al. ( | 2021 | Korean J Radiol | 42 AD, 85 MCI | 45 healthy | Inbrain | r, 0.72–96 | ||||
| Kim et al. ( | 2021 | Sci Rep | 26 AD | 55 MCI, | Hippocampus | 54 | 97 | 0.78 | ||
AD = Alzheimer’s disease, AUC = area under the curve, ICC = intraclass correlation, MCI = mild cognitive impairment, MTL = medial temporal lobe, SCI = subjective cognitive impairment
Artificial Intelligence-Augmented Automated Brain Volumetry and Classification Software in the Republic of Korea
| Year | Korean MFDS | Software | Company | Key Features | Main Techniques | Data | Evidence |
|---|---|---|---|---|---|---|---|
| 2017-04 | Grade 2 | Inbrain Morph | MIDAS IT | Brain volumetry | CNN (mainly for skull stripping and white matter segmentation) | Samsung Medical Center (3000 >) | Korean J Radiol 2021 ( |
| 2018-08 | Grade 2 | Neuro I | NEUROZEN | Brain volumetry | Atlas-based | National Research Center for Dementia (2000 >) | |
| 2019-06 | Grade 2 | DeepBrainTM | VUNO | Brain volumetry | CNN (2.5D HighResNet) + XGBoost | Asan Medical Center | AJNR Am J Neuroradiol 2020 ( |
| 2019-08 | Grade 2 | ASTROSCAN | JLK | Brain volumetry | 3D U-Net | ||
| 2020-12 | Grade 3 | DeepBrain AD | VUNO | Risk classification of AD | CNN (Inception v4) | Seoul National University Bundang Hospital | Sci Rep 2020 ( |
| 2021-03 | Grade 3 | Neurophet AQUA 2.0 | Neurophet | Brain volumetry | U-Net++ (ResNet with split-attention) | Yeouido St. Mary's Hospital, ST. Vincent's Hospital | Sci Rep 2020 ( |
AD = Alzheimer’s disease, CNN = convolutional neural network, MFDS = Ministry of Food and Drug Safety, NA = nonavailable
Artificial Intelligence-Augmented Automated Brain Volumetry or Classification Software from Global Companies (Except the Republic of Korea)
| Software | Company | Country | Key Features | Main Techniques | CE Mark | US FDA | |
|---|---|---|---|---|---|---|---|
| AIRAscore structure | AIRAmed | Germany | Brain volumetry | NA | Class I | ||
| Diadem | Brainminer | UK | Brain volumetry | Multi-atlas/MRF | Class I | ||
| Neuroreader® | Brainreader | Denmark | Brain volumetry | Multi-atlas | Class I | Class II | |
| cNeuro® cDSI | Combinostics | Finland | Clinical decision support in dementia | MRF (graph-cut)/EM | Class IIa | Class II | |
| NeuroQuant® | Cortechs.ai | Netherland | Brain volumetry | Multi-atlas (dynamic atlas) | Class IIa | Class II | |
| THINQ | CorticoMetrics | US | Brain volumetry | Atlas-based/MRF (FreeSurfer) | Class II | ||
| icobrain DM | icometrix | Belgium | Brain volumetry | Multi-atlas | Class I | Class II | |
| mdbrain | Mediaire | Germany | Brain volumetry | U-Net (ensemble) | Class I | ||
| Pixyl.Neuro.BV | Pixyl | France | Brain volumetry | MRF/EM | Class II | ||
| Quantib® ND | Quantib | Netherland | Brain volumetry | NA | Class IIa | Class II | |
| Neurocloud VOL | Qubiotech | Spain | Brain volumetry | NA | Class I | ||
| Brain atrophy screening | Quibim | Spain | Brain volumetry | CNN (2.5 D) | Class II | ||
| QyScore | Qynapse | France | Brain volumetry | NA | Class II | Class II | |
| AI-Rad Companion Brain MR | Siemens Healthineers | Germany | Brain volumetry | MRF/EM (tissue) | Class IIa | Class II | |
| Multi-atlas (region) | |||||||
CE = Conformite Europeenne, CNN = convolutional neural network, EM = expectation-maximization algorithm, MRF = markov random field, NA = nonavailable