Radiomics and Artificial Intelligence for the Diagnosis and Monitoring of Alzheimer's Disease: A Systematic Review of Studies in the Field

Article Radiology, Nuclear Medicine & Medical Imaging

Radiomics for Survival Risk Stratification of Clinical and Pathologic Stage IA Pure-Solid Non-Small Cell Lung Cancer

Tingting Wang et al.

Summary: In patients with clinical stage and pathologic stage IA pure-solid non-small cell lung cancer, using a radiomics signature with multiregional features helps stratify the survival risk.

RADIOLOGY (2022)

添加到收藏夹

Article Geriatrics & Gerontology

Radiomic Features of the Hippocampus for Diagnosing Early-Onset and Late-Onset Alzheimer's Disease

Yang Du et al.

Summary: This study aimed to differentiate between EOAD and LOAD using hippocampal radiomic features, providing valuable insights into potential biomarkers for early and late-onset Alzheimer's disease.

FRONTIERS IN AGING NEUROSCIENCE (2022)

添加到收藏夹

Article Medicine, General & Internal

Using Deep Learning Radiomics to Distinguish Cognitively Normal Adults at Risk of Alzheimer's Disease From Normal Control: An Exploratory Study Based on Structural MRI

Jiehui Jiang et al.

Summary: In this study, a novel deep learning radiomics (DLR) method was proposed to distinguish cognitively normal adults at risk of Alzheimer's disease (AD) from normal control based on T1-weighted structural MRI images. The DLR method achieved better classification performance than other models, providing potential clinical value in distinguishing preAD from NC.

FRONTIERS IN MEDICINE (2022)

添加到收藏夹

Article Geriatrics & Gerontology

Investigation of Underlying Association Between Whole Brain Regions and Alzheimer's Disease: A Research Based on an Artificial Intelligence Model

Shui Liu et al.

Summary: In this study, we analyzed the radiomic features of specific brain subregions that were most related to Alzheimer's disease (AD) using structural magnetic resonance imaging (sMRI) data. By building a machine learning model, we identified the hippocampus, the inferior parietal lobe, the precuneus, and the lateral occipital gyrus as the most relevant subregions to AD, which exhibited specific radiomic features. Ultimately, our model constructed using the logistic regression (LR) algorithm achieved high accuracy, providing a potential biomarker for predicting AD.

FRONTIERS IN AGING NEUROSCIENCE (2022)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

How segmentation methods affect hippocampal radiomic feature accuracy in Alzheimer's disease analysis?

Qiang Zheng et al.

Summary: This study investigates how different hippocampal segmentation methods affect the accuracy of hippocampal radiomic features (HRFs) in Alzheimer's disease (AD) analysis. The results show that HRFs exhibit high consistency across different segmentation methods and the best performance in AD classification is obtained when HRFs are extracted using the naive majority voting method with a more sufficient segmentation and relatively low hippocampal segmentation accuracy.

EUROPEAN RADIOLOGY (2022)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

Radiomics in predicting treatment response in non-small-cell lung cancer: current status, challenges and future perspectives

Madhurima R. Chetan et al.

Summary: The review summarized the current status of radiomics research in predicting treatment response in non-small-cell lung cancer, indicating low quality, lack of reproducibility, and limited clinical evaluation. Efforts towards standardization and collaboration are necessary to identify reproducible radiomic predictors of response. Promising radiomic models need external validation and evaluation within the clinical pathway before being implemented for personalized treatment in NSCLC patients.

EUROPEAN RADIOLOGY (2021)

添加到收藏夹

Review Clinical Neurology

Early-Onset Alzheimer's Disease: What Is Missing in Research?

Temitope Ayodele et al.

Summary: EOAD, occurring before age 65, is less studied than LOAD despite its aggressive disease progression. Only a fraction of the genetic variation has been identified, leaving the molecular mechanisms unclear. Further research is needed to identify potential targets for disease screening, diagnosis, prevention, and treatment.

CURRENT NEUROLOGY AND NEUROSCIENCE REPORTS (2021)

添加到收藏夹

Article Radiology, Nuclear Medicine & Medical Imaging

A CT-based Radiomics Signature Is Associated with Response to Immune Checkpoint Inhibitors in Advanced Solid Tumors

Marta Ligero et al.

Summary: A pretreatment CT-based radiomics signature is developed and validated to predict response to immune checkpoint inhibitors in advanced solid tumors. The radiomics signature is associated with clinical response to immune checkpoint inhibitors, potentially reflecting the tumor immunophenotype.

RADIOLOGY (2021)

添加到收藏夹

Article Neuroimaging

Comprehensive classification models based on amygdala radiomic features for Alzheimer's disease and mild cognitive impairment

Qi Feng et al.

Summary: This study established and validated classification models based on amygdala radiomic features for AD and aMCI, showing promising classification performances with good accuracy and AUC values between AD and aMCI, suggesting these features might serve as early biomarkers for detecting changes in microstructural brain tissue.

BRAIN IMAGING AND BEHAVIOR (2021)

添加到收藏夹

Review Radiology, Nuclear Medicine & Medical Imaging

Structural and functional radiomics for lung cancer

Guangyao Wu et al.

Summary: Radiomics plays a crucial role in the detection, diagnosis, and prediction of lung cancer, covering pathology, gene mutation, treatment strategy, follow-up, treatment response evaluation, and prognosis. However, challenges such as limited large datasets, methodology standardization, the black-box nature of deep learning, and reproducibility need to be addressed for the clinical implementation of radiomics. Future directions include developing a safer and more efficient model training mode, merging multi-modality images, and combining multi-discipline or multi-omics to form Medomics.

EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING (2021)

添加到收藏夹

Review Radiology, Nuclear Medicine & Medical Imaging

Delta radiomics: a systematic review

Valerio Nardone et al.

Summary: Background radiomics and delta radiomics play important roles in medical imaging, correlating with biological features and clinical endpoints, as well as analyzing feature variations. Studies were categorized by disease types, with an overall low quality assessment, leading to potentially inconsistent and unstable conclusions in the current literature. Prospective and multicenter studies are necessary for further clinical validation of delta radiomics approaches.

RADIOLOGIA MEDICA (2021)

添加到收藏夹

Review Radiology, Nuclear Medicine & Medical Imaging

AI applications to medical images: From machine learning to deep learning

Isabella Castiglioni et al.

Summary: This review focuses on the challenges associated with developing AI applications as clinical decision support systems in the context of biomedical research and healthcare services. It outlines the architectural characteristics of machine learning (ML)/radiomics and deep learning (DL) approaches, and discusses specific challenges such as data curation, interpretability of AI models, and the choice between ML and DL for medical imaging applications.

PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS (2021)

添加到收藏夹

Article Neurosciences

Quantitative Radiomic Features as New Biomarkers for Alzheimer's Disease: An Amyloid PET Study

Yanhui Ding et al.

Summary: The study revealed that the radiomic features of A beta PET images have high accuracy in distinguishing AD from normal controls and predicting the conversion of mild cognitive impairment to AD, providing new biomarkers for clinical applications in AD/MCI.

CEREBRAL CORTEX (2021)

添加到收藏夹

Article Multidisciplinary Sciences

Predicting amyloid positivity in patients with mild cognitive impairment using a radiomics approach

Jun Pyo Kim et al.

Summary: Predicting amyloid positivity in MCI patients is crucial, and in this study, radiomics features derived from structural MRI were found to have predictive value. Combining T1 and T2 FLAIR radiomics models with baseline features resulted in the best predictive performance for amyloid positivity.

SCIENTIFIC REPORTS (2021)

添加到收藏夹

Review Medicine, General & Internal

The PRISMA 2020 statement: an updated guideline for reporting systematic reviews

Matthew J. Page et al.

Summary: The PRISMA statement was designed to help systematic reviewers transparently report the purpose, methods, and findings of their reviews. The updated PRISMA 2020 statement includes new reporting guidance, a 27-item checklist, an abstract checklist, and revised flow diagrams for reviews.

BMJ-BRITISH MEDICAL JOURNAL (2021)

添加到收藏夹

Review Gastroenterology & Hepatology

Systematic review: Radiomics for the diagnosis and prognosis of hepatocellular carcinoma

Emily Harding-Theobald et al.

Summary: Studies have shown that radiomic features have good discriminatory performance in differentiating HCC from other solid lesions, predicting microvascular invasion, early recurrence after hepatectomy, and prognosis after therapies. However, the overall quality of the included studies was low due to common deficiencies in internal and external validation, standardized imaging segmentation, and comparison to a gold standard. Therefore, standardization of protocols and outcome measurements, sharing of algorithms and analytic methods, and external validation are necessary before widespread application of radiomics in HCC diagnosis and prognosis in clinical practice.

ALIMENTARY PHARMACOLOGY & THERAPEUTICS (2021)

添加到收藏夹

Article Geriatrics & Gerontology

Radiomics approach in the neurodegenerative brain

Christian Salvatore et al.

Summary: Radiomics is a method that converts digital medical images into mineable data, which has the potential to uncover disease characteristics and has been widely applied in oncology and genetic fields. Its application in neurodegenerative disorders is relatively new.

AGING CLINICAL AND EXPERIMENTAL RESEARCH (2021)

添加到收藏夹

Review Biochemistry & Molecular Biology

Application of radiomics and machine learning in head and neck cancers

Zhouying Peng et al.

Summary: Radiomics, a new omics field, extracts high-throughput quantitative data from digital medical images, which integrates with clinical data, pathology, and molecular biomarkers to reveal tumor development more closely. Machine learning has a unique advantage in processing radiomic data, enabling the construction of models that accurately reflect tumor development for personalized and sequential patient treatment. This has broad application prospects in head and neck cancers.

INTERNATIONAL JOURNAL OF BIOLOGICAL SCIENCES (2021)

添加到收藏夹

Review Clinical Neurology