-
Translational Vision Science &... Feb 2024Retinal images contain rich biomarker information for neurodegenerative disease. Recently, deep learning models have been used for automated neurodegenerative disease...
PURPOSE
Retinal images contain rich biomarker information for neurodegenerative disease. Recently, deep learning models have been used for automated neurodegenerative disease diagnosis and risk prediction using retinal images with good results.
METHODS
In this review, we systematically report studies with datasets of retinal images from patients with neurodegenerative diseases, including Alzheimer's disease, Huntington's disease, Parkinson's disease, amyotrophic lateral sclerosis, and others. We also review and characterize the models in the current literature which have been used for classification, regression, or segmentation problems using retinal images in patients with neurodegenerative diseases.
RESULTS
Our review found several existing datasets and models with various imaging modalities primarily in patients with Alzheimer's disease, with most datasets on the order of tens to a few hundred images. We found limited data available for the other neurodegenerative diseases. Although cross-sectional imaging data for Alzheimer's disease is becoming more abundant, datasets with longitudinal imaging of any disease are lacking.
CONCLUSIONS
The use of bilateral and multimodal imaging together with metadata seems to improve model performance, thus multimodal bilateral image datasets with patient metadata are needed. We identified several deep learning tools that have been useful in this context including feature extraction algorithms specifically for retinal images, retinal image preprocessing techniques, transfer learning, feature fusion, and attention mapping. Importantly, we also consider the limitations common to these models in real-world clinical applications.
TRANSLATIONAL RELEVANCE
This systematic review evaluates the deep learning models and retinal features relevant in the evaluation of retinal images of patients with neurodegenerative disease.
Topics: Humans; Algorithms; Alzheimer Disease; Deep Learning; Machine Learning; Neurodegenerative Diseases; Datasets as Topic; Retina
PubMed: 38381447
DOI: 10.1167/tvst.13.2.16 -
Journal of Cranio-maxillo-facial... Apr 2024Imaging with bone scans plays an important role in the diagnostic path of patients with unilateral condylar hyperactivity or unilateral condylar hyperplasia (UCH). The... (Meta-Analysis)
Meta-Analysis Review
Imaging with bone scans plays an important role in the diagnostic path of patients with unilateral condylar hyperactivity or unilateral condylar hyperplasia (UCH). The aim of this study is to perform a systematic review of the diagnostic performance of the bone SPECT and SPECT/CT scan for the diagnosis of UCH. PubMed, SCOPUS and EMBASE were searched electronically to identify diagnostic accuracy studies that assessed the diagnostic value of bone SPECT and SPECT/CT for the diagnosis of UCH, Meta-analyses were performed with Metadisc 1.4 and 2.0. A total of 14 studies, with a total number of 887 patients, were included in the qualitative analysis and 11 studies qualified for meta-analyses. The pooled sensitivity and specificity for the SPECT scan were 0.814 (95 % CI: 0.639-0.915) and 0.774 (95 % CI: 0.655-0.861), for the SPECT/CT scan these were 0.818 (95 % CI: 0.749-0.874) and 0.901 (95 % CI: 0.840-0.945). The summary receiver operating characteristics of the SPECT scan showed an area under the curve of 0.847 (95 % CI: 0.722-0.972) and that of the SPECT/CT scan was 0.928 (95 % CI: 0.876-0.980). CONCLUSION: Both bone SPECT scan and SPECT/CT scan provide a high diagnostic accuracy for UCH. The added value of the SPECT/CT scan is questionable and given the potential disadvantages of the SPECT/CT scan, including the increased radiation dose and costs, the diagnostic modality of first choice in patients with UCH should be a SPECT scan.
Topics: Humans; Hyperplasia; Mandibular Condyle; Tomography, Emission-Computed, Single-Photon; Radionuclide Imaging; Single Photon Emission Computed Tomography Computed Tomography; Stomatognathic Diseases; Bone Diseases
PubMed: 38378369
DOI: 10.1016/j.jcms.2024.01.013 -
European Journal of Radiology Mar 2024X-ray imaging plays a crucial role in diagnostic medicine. Yet, a significant portion of the global population lacks access to this essential technology due to a... (Review)
Review
X-ray imaging plays a crucial role in diagnostic medicine. Yet, a significant portion of the global population lacks access to this essential technology due to a shortage of trained radiologists. Eye-tracking data and deep learning models can enhance X-ray analysis by mapping expert focus areas, guiding automated anomaly detection, optimizing workflow efficiency, and bolstering training methods for novice radiologists. However, the literature shows contradictory results regarding the usefulness of eye-tracking data in deep-learning architectures for abnormality detection. We argue that these discrepancies between studies in the literature are due to (a) the way eye-tracking data is (or is not) processed, (b) the types of deep learning architectures chosen, and (c) the type of application that these architectures will have. We conducted a systematic literature review using PRISMA to address these contradicting results. We analyzed 60 studies that incorporated eye-tracking data in a deep-learning approach for different application goals in radiology. We performed a comparative analysis to understand if eye gaze data contains feature maps that can be useful under a deep learning approach and whether they can promote more interpretable predictions. To the best of our knowledge, this is the first survey in the area that performs a thorough investigation of eye gaze data processing techniques and their impacts in different deep learning architectures for applications such as error detection, classification, object detection, expertise level analysis, fatigue estimation and human attention prediction in medical imaging data. Our analysis resulted in two main contributions: (1) taxonomy that first divides the literature by task, enabling us to analyze the value eye movement can bring for each case and build guidelines regarding architectures and gaze processing techniques adequate for each application, and (2) an overall analysis of how eye gaze data can promote explainability in radiology.
Topics: Humans; Fixation, Ocular; Deep Learning; Radiography; Radiology; Eye Movements
PubMed: 38340426
DOI: 10.1016/j.ejrad.2024.111341 -
Ageing Research Reviews Feb 2024Positron emission tomography (PET) with radiotracers that bind to synaptic vesicle glycoprotein 2 A (SV2A) enables quantification of synaptic density in the living...
Positron emission tomography (PET) with radiotracers that bind to synaptic vesicle glycoprotein 2 A (SV2A) enables quantification of synaptic density in the living human brain. Assessing the regional distribution and severity of synaptic density loss will contribute to our understanding of the pathological processes that precede atrophy in neurodegeneration. In this systematic review, we provide a discussion of in vivo SV2A PET imaging research for quantitative assessment of synaptic density in various dementia conditions: amnestic Mild Cognitive Impairment and Alzheimer's disease, Frontotemporal dementia, Progressive supranuclear palsy and Corticobasal degeneration, Parkinson's disease and Dementia with Lewy bodies, Huntington's disease, and Spinocerebellar Ataxia. We discuss the main findings concerning group differences and clinical-cognitive correlations, and explore relations between SV2A PET and other markers of pathology. Additionally, we touch upon synaptic density in healthy ageing and outcomes of radiotracer validation studies. Studies were identified on PubMed and Embase between 2018 and 2023; last searched on the 3rd of July 2023. A total of 36 studies were included, comprising 5 on normal ageing, 21 clinical studies, and 10 validation studies. Extracted study characteristics were participant details, methodological aspects, and critical findings. In summary, the small but growing literature on in vivo SV2A PET has revealed different spatial patterns of synaptic density loss among various neurodegenerative disorders that correlate with cognitive functioning, supporting the potential role of SV2A PET imaging for differential diagnosis. SV2A PET imaging shows tremendous capability to provide novel insights into the aetiology of neurodegenerative disorders and great promise as a biomarker for synaptic density reduction. Novel directions for future synaptic density research are proposed, including (a) longitudinal imaging in larger patient cohorts of preclinical dementias, (b) multi-modal mapping of synaptic density loss onto other pathological processes, and (c) monitoring therapeutic responses and assessing drug efficacy in clinical trials.
Topics: Humans; Alzheimer Disease; Brain; Cognitive Dysfunction; Neurodegenerative Diseases; Positron-Emission Tomography
PubMed: 38266660
DOI: 10.1016/j.arr.2024.102197 -
Radiotherapy and Oncology : Journal of... Mar 2024Radiation therapy is used frequently for patients with prostate cancer. Dose escalation to intraprostatic lesions (IPLs) has been shown to improve oncologic outcomes,... (Meta-Analysis)
Meta-Analysis
Using multiparametric Magnetic Resonance Imaging and Prostate Specific Membrane Antigen Positron Emission Tomography to detect and delineate the gross tumour volume of intraprostatic lesions - A systematic review and meta-analysis.
BACKGROUND AND PURPOSE
Radiation therapy is used frequently for patients with prostate cancer. Dose escalation to intraprostatic lesions (IPLs) has been shown to improve oncologic outcomes, without increasing toxicity. Both multiparametric MRI (mpMRI) and PSMA PET can be used to identify IPLs.
MATERIALS AND METHODS
A systematic review was conducted to determine the ability of mpMRI, PSMA PET and their combination to detect IPLs prior to radical prostatectomy (RP) as correlated with the histology. Trials included patients that had mpMRI, PSMA PET, or both, prior to RP. The quality of the histopathological-radiological co-registration was assessed as high or low for each study. Recorded outcomes include sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC). A meta-analysis was conducted using a bivariate model to determine the pooled sensitivity and specificity for each imaging modality. This systematic review was registered through PROSPERO (CRD42023389092).
RESULTS
Altogether, 42 studies were included in the systematic review. Of these, 20 could be included in the meta-analysis. The pooled sensitivity (95 % CI), specificity (95 % CI) and AUROC for mpMRI (n = 13 studies) were 64.7 % (50.2 % - 76.9 %), 86.4 % (79.7 % - 91.1 %), and 0.852; the pooled outcomes for PSMA PET (n = 12) were 75.7 % (64.0 % - 84.5 %), 87.1 % (80.2 % - 91.9 %), and 0.889; for their combination (n = 5), the pooled outcomes were 70.3 % (64.1 % - 75.9 %), 81.9 % (71.9 % - 88.8 %), and 0.796. When reviewing studies with a high-quality histopathological-radiological co-registration, IPL delineation recommendations varied by study and the imaging modality used.
CONCLUSION
All of mpMRI, PSMA PET or their combination were found to have very good diagnostic outcomes for detecting IPLs. Recommendations for delineating IPLs varied based on the imaging modalities used and between research groups. Consensus guidelines for IPL delineation would help with creating consistency for focal boost radiation treatments in future studies.
Topics: Male; Humans; Multiparametric Magnetic Resonance Imaging; Prostate; Tumor Burden; Gallium Radioisotopes; Positron Emission Tomography Computed Tomography; Prostatic Neoplasms; Positron-Emission Tomography; Magnetic Resonance Imaging
PubMed: 38262815
DOI: 10.1016/j.radonc.2023.110070 -
European Journal of Radiology Feb 2024To summarize the underlying biological correlation of prognostic radiomics and deep learning signatures in patients with lung cancer and evaluate the quality of... (Review)
Review
OBJECTIVES
To summarize the underlying biological correlation of prognostic radiomics and deep learning signatures in patients with lung cancer and evaluate the quality of available studies.
METHODS
This study examined databases including the PubMed, Embase, Web of Science Core Collection, and Cochrane Library, for studies that elaborated on the underlying biological correlation with prognostic radiomics and deep learning signatures based on CT or PET/CT for predicting the prognosis in patients with lung cancer. Information about the patient and radiogenomic analyses was extracted for the included studies. The Radiomics Quality Score (RQS) and the Prediction Model Risk of Bias Assessment Tool were used to assess the quality of these studies.
RESULTS
Twelve studies were included with 7,338 patients from 2014 to 2022. All studies except for one were retrospective. Supervised machine learning was adopted in six studies, and the remaining used unsupervised machine learning methods. Gene sequencing and histopathological data were analyzed by 83.33% and 16.67% of the included studies, respectively. Gene set enrichment analysis and correlation analysis were most used to explore the biological meaning of prognostic signatures. The median RQS for supervised learning articles was 13.5 (range 12-19) and 7.0 (range 5-14) for unsupervised learning articles. The studies included in this report were assessed to have high risk of bias overall.
CONCLUSION
The biological basis for the interpretability of data-driven models mainly focused on genomics and histopathological factors, and it may improve the prognosis of lung cancer with more proper biological interpretation in the future.
Topics: Humans; Prognosis; Lung Neoplasms; Deep Learning; Positron Emission Tomography Computed Tomography; Radiomics; Retrospective Studies
PubMed: 38244306
DOI: 10.1016/j.ejrad.2024.111314 -
European Journal of Nuclear Medicine... Jul 2024This study aimed to evaluate the level of evidence of expert recommendations and guidelines for clinical indications and procedurals in hybrid nuclear cardiovascular... (Review)
Review
OBJECTIVES
This study aimed to evaluate the level of evidence of expert recommendations and guidelines for clinical indications and procedurals in hybrid nuclear cardiovascular imaging.
METHODS
From inception to August 2023, a PubMed literature analysis of the latest version of guidelines for clinical hybrid cardiovascular imaging techniques including SPECT(/CT), PET(/CT), and PET(/MRI) was performed in two categories: (1) for clinical indications for all-in primary diagnosis; subgroup in prognosis and therapy evaluation; and for (2) imaging procedurals. We surveyed to what degree these followed a standard methodology to collect the data and provide levels of evidence, and for which topic systematic review evidence was executed.
RESULTS
A total of 76 guidelines, published between 2013 and 2023, were included. The evidence of guidelines was based on systematic reviews in 7.9% of cases, non-systematic reviews in 47.4% of cases, a mix of systematic and non-systematic reviews in 19.7%, and 25% of guidelines did not report any evidence. Search strategy was reported in 36.8% of cases. Strengths of recommendation were clearly reported in 25% of guidelines. The notion of external review was explicitly reported in 23.7% of cases. Finally, the support of a methodologist was reported in 11.8% of the included guidelines.
CONCLUSION
The use of evidence procedures for developing for evidence-based cardiovascular hybrid imaging recommendations and guidelines is currently suboptimal, highlighting the need for more standardized methodological procedures.
Topics: Humans; Practice Guidelines as Topic; Multimodal Imaging; Evidence-Based Medicine; Cardiovascular Diseases; Nuclear Medicine
PubMed: 38221570
DOI: 10.1007/s00259-024-06597-x -
Frontiers in Neurology 2023Parkinson's disease (PD) is a neurodegenerative disease with high incidence rate. Resting state functional magnetic resonance imaging (rs-fMRI), as a widely used method...
BACKGROUND
Parkinson's disease (PD) is a neurodegenerative disease with high incidence rate. Resting state functional magnetic resonance imaging (rs-fMRI), as a widely used method for studying neurodegenerative diseases, has not yet been combined with two important indicators, amplitude low-frequency fluctuation (ALFF) and cerebral blood flow (CBF), for standardized analysis of PD.
METHODS
In this study, we used seed-based d-mapping and permutation of subject images (SDM-PSI) software to investigate the changes in ALFF and CBF of PD patients. After obtaining the regions of PD with changes in ALFF or CBF, we conducted a multimodal analysis to identify brain regions where ALFF and CBF changed together or could not synchronize.
RESULTS
The final study included 31 eligible trials with 37 data sets. The main analysis results showed that the ALFF of the left striatum and left anterior thalamic projection decreased in PD patients, while the CBF of the right superior frontal gyrus decreased. However, the results of multimodal analysis suggested that there were no statistically significant brain regions. In addition, the decrease of ALFF in the left striatum and the decrease of CBF in the right superior frontal gyrus was correlated with the decrease in clinical cognitive scores.
CONCLUSION
PD patients had a series of spontaneous brain activity abnormalities, mainly involving brain regions related to the striatum-thalamic-cortex circuit, and related to the clinical manifestations of PD. Among them, the left striatum and right superior frontal gyrus are more closely related to cognition.
SYSTEMATIC REVIEW REGISTRATION
https://www.crd.york.ac.uk/ PROSPERO (CRD42023390914).
PubMed: 38162449
DOI: 10.3389/fneur.2023.1289934 -
Journal of Imaging Dec 2023Due to the importance of correct and timely diagnosis of bone metastases in advanced breast cancer (BrC), we performed a meta-analysis evaluating the diagnostic accuracy... (Review)
Review
Due to the importance of correct and timely diagnosis of bone metastases in advanced breast cancer (BrC), we performed a meta-analysis evaluating the diagnostic accuracy of [F]FDG, or Na[F]F PET, PET(/CT), and (/MRI) versus [Tc]Tc-diphosphonates bone scintigraphy (BS). The PubMed, Embase, Scopus, and Scholar electronic databases were searched. The results of the selected studies were analyzed using pooled sensitivity and specificity, diagnostic odds ratio (DOR), positive-negative likelihood ratio (LR-LR), and summary receiver-operating characteristic (SROC) curves. Eleven studies including 753 BrC patients were included in the meta-analysis. The patient-based pooled values of sensitivity, specificity, and area under the SROC curve (AUC) for BS (with 95% confidence interval values) were 90% (86-93), 91% (87-94), and 0.93, respectively. These indices for [F]FDG PET(/CT) were 92% (88-95), 99% (96-100), and 0.99, respectively, and for Na[F]F PET(/CT) were 96% (90-99), 81% (72-88), and 0.99, respectively. BS has good diagnostic performance in detecting BrC bone metastases. However, due to the higher and balanced sensitivity and specificity of [F]FDG PET(/CT) compared to BS and Na[F]F PET(/CT), and its advantage in evaluating extra-skeletal lesions, [F]FDG PET(/CT) should be the preferred multimodal imaging method for evaluating bone metastases of BrC, if available.
PubMed: 38132692
DOI: 10.3390/jimaging9120274 -
Cells Dec 2023Progressive supranuclear palsy (PSP) is a neurodegenerative disease characterized by four-repeat tau deposition in various cell types and anatomical regions, and can... (Review)
Review
Progressive supranuclear palsy (PSP) is a neurodegenerative disease characterized by four-repeat tau deposition in various cell types and anatomical regions, and can manifest as several clinical phenotypes, including the most common phenotype, Richardson's syndrome. The limited availability of biomarkers for PSP relates to the overlap of clinical features with other neurodegenerative disorders, but identification of a growing number of biomarkers from imaging is underway. One way to increase the reliability of imaging biomarkers is to combine different modalities for multimodal imaging. This review aimed to provide an overview of the current state of PSP hybrid imaging by combinations of positron emission tomography (PET) and magnetic resonance imaging (MRI). Specifically, combined PET and MRI studies in PSP highlight the potential of [18F]AV-1451 to detect tau, but also the challenge in differentiating PSP from other neurodegenerative diseases. Studies over the last years showed a reduced synaptic density in [11C]UCB-J PET, linked [11C]PK11195 and [18F]AV-1451 markers to disease progression, and suggested the potential role of [18F]RO948 PET for identifying tau pathology in subcortical regions. The integration of quantitative global and regional gray matter analysis by MRI may further guide the assessment of reduced cortical thickness or volume alterations, and diffusion MRI could provide insight into microstructural changes and structural connectivity in PSP. Challenges in radiopharmaceutical biomarkers and hybrid imaging require further research targeting markers for comprehensive PSP diagnosis.
Topics: Humans; Supranuclear Palsy, Progressive; Radiopharmaceuticals; Neurodegenerative Diseases; Reproducibility of Results; Multimodal Imaging; Biomarkers
PubMed: 38132096
DOI: 10.3390/cells12242776