Visual interpretation of [18F]Florbetaben PET supported by deep learning–based estimation of amyloid burden

New objective simple evaluation methods of amyloid PET/CT using whole-brain histogram and Top20%-Map

OBJECTIVE

This study aims to assess the utility of newly developed objective methods for the evaluation of intracranial abnormal amyloid deposition using PET/CT histogram without use of cortical ROI analyses.

METHODS

Twenty-five healthy volunteers (HV) and 38 patients with diagnosed or suspected dementia who had undergone 18F-FPYBF-2 PET/CT were retrospectively included in this study. Out of them, 11C-PiB PET/CT had been also performed in 13 subjects. In addition to the conventional methods, namely visual judgment and quantitative analyses using composed standardized uptake value ratio (comSUVR), the PET images were also evaluated by the following new parameters: the skewness and the mode-to-mean ratio (MMR) obtained from the histogram of the brain parenchyma; Top20%-map highlights the areas with high tracer accumulation occupying 20% volume of the total brain parenchymal on the individual's CT images. We evaluated the utility of the new methods using histogram compared with the visual assessment and comSUVR. The results of these new methods between 18F-FPYBF-2 and 11C-PiB were also compared in 13 subjects.

RESULTS

In visual analysis, 32, 9, and 22 subjects showed negative, border, and positive results, and composed SUVR in each group were 1.11 ± 0.06, 1.20 ± 0.13, and 1.48 ± 0.18 (p < 0.0001), respectively. Visually positive subjects showed significantly low skewness and high MMR (p < 0.0001), and the Top20%-Map showed the presence or absence of abnormal deposits clearly. In comparison between the two tracers, visual evaluation was all consistent, and the ComSUVR, the skewness, the MMR showed significant good correlation. The Top20%-Maps showed similar pattern.

CONCLUSIONS

Our new methods using the histogram of the brain parenchymal accumulation are simple and suitable for clinical practice of amyloid PET, and Top20%-Map on the individual's brain CT can be of great help for the visual assessment.

AmyloidPETNet: Classification of Amyloid Positivity in Brain PET Imaging Using End-to-End Deep Learning

Background Visual assessment of amyloid PET scans relies on the availability of radiologist expertise, whereas quantification of amyloid burden typically involves MRI for processing and analysis, which can be computationally expensive. Purpose To develop a deep learning model to classify minimally processed brain PET scans as amyloid positive or negative, evaluate its performance on independent data sets and different tracers, and compare it with human visual reads. Materials and Methods This retrospective study used 8476 PET scans (6722 patients) obtained from late 2004 to early 2023 that were analyzed across five different data sets. A deep learning model, AmyloidPETNet, was trained on 1538 scans from 766 patients, validated on 205 scans from 95 patients, and internally tested on 184 scans from 95 patients in the Alzheimer's Disease Neuroimaging Initiative (ADNI) fluorine 18 (18F) florbetapir (FBP) data set. It was tested on ADNI scans using different tracers and scans from independent data sets. Scan amyloid positivity was based on mean cortical standardized uptake value ratio cutoffs. To compare with model performance, each scan from both the Centiloid Project and a subset of the Anti-Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4) study were visually interpreted with a confidence level (low, intermediate, high) of amyloid positivity/negativity. The area under the receiver operating characteristic curve (AUC) and other performance metrics were calculated, and Cohen κ was used to measure physician-model agreement. Results The model achieved an AUC of 0.97 (95% CI: 0.95, 0.99) on test ADNI 18F-FBP scans, which generalized well to 18F-FBP scans from the Open Access Series of Imaging Studies (AUC, 0.95; 95% CI: 0.93, 0.97) and the A4 study (AUC, 0.98; 95% CI: 0.98, 0.98). Model performance was high when applied to data sets with different tracers (AUC ≥ 0.97). Other performance metrics provided converging evidence. Physician-model agreement ranged from fair (Cohen κ = 0.39; 95% CI: 0.16, 0.60) on a sample of mostly equivocal cases from the A4 study to almost perfect (Cohen κ = 0.93; 95% CI: 0.86, 1.0) on the Centiloid Project. Conclusion The developed model was capable of automatically and accurately classifying brain PET scans as amyloid positive or negative without relying on experienced readers or requiring structural MRI. Clinical trial registration no. NCT00106899 © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Bryan and Forghani in this issue.

Deep Learning-Driven Estimation of Centiloid Scales from Amyloid PET Images with 11C-PiB and 18F-Labeled Tracers in Alzheimer's Disease

BACKGROUND

Standard methods for deriving Centiloid scales from amyloid PET images are time-consuming and require considerable expert knowledge. We aimed to develop a deep learning method of automating Centiloid scale calculations from amyloid PET images with 11C-Pittsburgh Compound-B (PiB) tracer and assess its applicability to 18F-labeled tracers without retraining.

METHODS

We trained models on 231 11C-PiB amyloid PET images using a 50-layer 3D ResNet architecture. The models predicted the Centiloid scale, and accuracy was assessed using mean absolute error (MAE), linear regression analysis, and Bland-Altman plots.

RESULTS

The MAEs for Alzheimer's disease (AD) and young controls (YC) were 8.54 and 2.61, respectively, using 11C-PiB, and 8.66 and 3.56, respectively, using 18F-NAV4694. The MAEs for AD and YC were higher with 18F-florbetaben (39.8 and 7.13, respectively) and 18F-florbetapir (40.5 and 12.4, respectively), and the error rate was moderate for 18F-flutemetamol (21.3 and 4.03, respectively). Linear regression yielded a slope of 1.00, intercept of 1.26, and R2 of 0.956, with a mean bias of -1.31 in the Centiloid scale prediction.

CONCLUSIONS

We propose a deep learning means of directly predicting the Centiloid scale from amyloid PET images in a native space. Transferring the model trained on 11C-PiB directly to 18F-NAV4694 without retraining was feasible.

Validation of quantitative assessment of florbetaben PET scans as an adjunct to the visual assessment across 15 software methods

PURPOSE

Amyloid positron emission tomography (PET) with [18F]florbetaben (FBB) is an established tool for detecting Aβ deposition in the brain in vivo based on visual assessment of PET scans. Quantitative measures are commonly used in the research context and allow continuous measurement of amyloid burden. The aim of this study was to demonstrate the robustness of FBB PET quantification.

METHODS

This is a retrospective analysis of FBB PET images from 589 subjects. PET scans were quantified with 15 analytical methods using nine software packages (MIMneuro, Hermes BRASS, Neurocloud, Neurology Toolkit, statistical parametric mapping (SPM8), PMOD Neuro, CapAIBL, non-negative matrix factorization (NMF), AmyloidIQ) that used several metrics to estimate Aβ load (SUVR, centiloid, amyloid load, and amyloid index). Six analytical methods reported centiloid (MIMneuro, standard centiloid, Neurology Toolkit, SPM8 (PET only), CapAIBL, NMF). All results were quality controlled.

RESULTS

The mean sensitivity, specificity, and accuracy were 96.1 ± 1.6%, 96.9 ± 1.0%, and 96.4 ± 1.1%, respectively, for all quantitative methods tested when compared to histopathology, where available. The mean percentage of agreement between binary quantitative assessment across all 15 methods and visual majority assessment was 92.4 ± 1.5%. Assessments of reliability, correlation analyses, and comparisons across software packages showed excellent performance and consistent results between analytical methods.

CONCLUSION

This study demonstrated that quantitative methods using both CE marked software and other widely available processing tools provided comparable results to visual assessments of FBB PET scans. Software quantification methods, such as centiloid analysis, can complement visual assessment of FBB PET images and could be used in the future for identification of early amyloid deposition, monitoring disease progression and treatment effectiveness.

Machine learning in the positron emission tomography imaging of Alzheimer's disease

The utilization of machine learning techniques in medicine has exponentially increased over the last decades due to innovations in computer processing, algorithm development, and access to big data. Applications of machine learning techniques to neuroimaging specifically have unveiled various hidden interactions, structures, and mechanisms related to various neurological disorders. One application of interest is the imaging of Alzheimer's disease, the most common cause of progressive dementia. The diagnoses of Alzheimer's disease, mild cognitive impairment, and preclinical Alzheimer's disease have been difficult. Molecular imaging, particularly via PET scans, holds tremendous value in the imaging of Alzheimer's disease. To date, many novel algorithms have been developed with great success that leverage machine learning in the context of Alzheimer's disease. This review article provides an overview of the diverse applications of machine learning to PET imaging of Alzheimer's disease.

International Nuclear Medicine Consensus on the Clinical Use of Amyloid Positron Emission Tomography in Alzheimer's Disease

Alzheimer's disease (AD) is the main cause of dementia, with its diagnosis and management remaining challenging. Amyloid positron emission tomography (PET) has become increasingly important in medical practice for patients with AD. To integrate and update previous guidelines in the field, a task group of experts of several disciplines from multiple countries was assembled, and they revised and approved the content related to the application of amyloid PET in the medical settings of cognitively impaired individuals, focusing on clinical scenarios, patient preparation, administered activities, as well as image acquisition, processing, interpretation and reporting. In addition, expert opinions, practices, and protocols of prominent research institutions performing research on amyloid PET of dementia are integrated. With the increasing availability of amyloid PET imaging, a complete and standard pipeline for the entire examination process is essential for clinical practice. This international consensus and practice guideline will help to promote proper clinical use of amyloid PET imaging in patients with AD.

Estimation of brain amyloid accumulation using deep learning in clinical [11C]PiB PET imaging

INTRODUCTION

Estimation of brain amyloid accumulation is valuable for evaluation of patients with cognitive impairment in both research and clinical routine. The development of high throughput and accurate strategies for the determination of amyloid status could be an important tool in patient selection for clinical trials and amyloid directed treatment. Here, we propose the use of deep learning to quantify amyloid accumulation using standardized uptake value ratio (SUVR) and classify amyloid status based on their PET images.

METHODS

A total of 1309 patients with cognitive impairment scanned with [¹¹C]PIB PET/CT or PET/MRI were included. Two convolutional neural networks (CNNs) for reading-based amyloid status and SUVR prediction were trained using 75% of the PET/CT data. The remaining PET/CT (n = 300) and all PET/MRI (n = 100) data was used for evaluation.

RESULTS

The prevalence of amyloid positive patients was 61%. The amyloid status classification model reproduced the expert reader's classification with 99% accuracy. There was a high correlation between reference and predicted SUVR (R² = 0.96). Both reference and predicted SUVR had an accuracy of 97% compared to expert classification when applying a predetermined SUVR threshold of 1.35 for binary classification of amyloid status.

CONCLUSION

The proposed CNN models reproduced both the expert classification and quantitative measure of amyloid accumulation in a large local dataset. This method has the potential to replace or simplify existing clinical routines and can facilitate fast and accurate classification well-suited for a high throughput pipeline.

Classification of Aβ State From Brain Amyloid PET Images Using Machine Learning Algorithm

Background and Purpose

Analyzing brain amyloid positron emission tomography (PET) images to access the occurrence of β-amyloid (Aβ) deposition in Alzheimer's patients requires much time and effort from physicians, while the variation of each interpreter may differ. For these reasons, a machine learning model was developed using a convolutional neural network (CNN) as an objective decision to classify the Aβ positive and Aβ negative status from brain amyloid PET images.

Methods

A total of 7,344 PET images of 144 subjects were used in this study. The 18F-florbetaben PET was administered to all participants, and the criteria for differentiating Aβ positive and Aβ negative state was based on brain amyloid plaque load score (BAPL) that depended on the visual assessment of PET images by the physicians. We applied the CNN algorithm trained in batches of 51 PET images per subject directory from 2 classes: Aβ positive and Aβ negative states, based on the BAPL scores.

Results

The binary classification of the model average performance matrices was evaluated after 40 epochs of three trials based on test datasets. The model accuracy for classifying Aβ positivity and Aβ negativity was (95.00±0.02) in the test dataset. The sensitivity and specificity were (96.00±0.02) and (94.00±0.02), respectively, with an area under the curve of (87.00±0.03).

Conclusions

Based on this study, the designed CNN model has the potential to be used clinically to screen amyloid PET images.

Automated semi-quantitative amyloid PET analysis technique without MR images for Alzheimer's disease

Objective

Although beta-amyloid (Aβ) positron emission tomography (PET) images are interpreted visually as positive or negative, approximately 10% are judged as equivocal in Alzheimer’s disease. Therefore, we aimed to develop an automated semi-quantitative analysis technique using ¹⁸F-flutemetamol PET images without anatomical images.

Methods

Overall, 136 cases of patients administered ¹⁸F-flutemetamol were enrolled. Of 136 cases, five PET images each with the highest and lowest values of standardized uptake value ratio (SUVr) of cerebral cortex-to-pons were used to create positive and negative templates. Using these templates, PET images of the remaining 126 cases were standardized, and SUVr images were produced with the pons as a reference region. The mean of SUVr values in the volume of interest delineated on the cerebral cortex was compared to those in the CortexID Suite (GE Healthcare). Furthermore, centiloid (CL) values were calculated for the 126 cases using data from the Centiloid Project (http://www.gaain.org/centiloid-project) and both templates. ¹⁸F-flutemetamol-PET was interpreted visually as positive/negative based on Aβ deposition in the cortex. However, the criterion "equivocal" was added for cases with focal or mild Aβ accumulation that were difficult to categorize. Optimal cutoff values of SUVr and CL maximizing sensitivity and specificity for Aβ detection were determined by receiver operating characteristic (ROC) analysis using the visual evaluation as a standard of truth.

Results

SUVr calculated by our method and CortexID were highly correlated (R² = 0.9657). The 126 PET images comprised 84 negative and 42 positive cases of Aβ deposition by visual evaluation, of which 11 and 10 were classified as equivocal, respectively. ROC analyses determined the optimal cutoff values, sensitivity, and specificity for SUVr as 0.544, 89.3%, and 92.9%, respectively, and for CL as 12.400, 94.0%, and 92.9%, respectively. Both semi-quantitative analyses showed that 12 and 9 of the 21 equivocal cases were negative and positive, respectively, under the optimal cutoff values.

Conclusions

This semi-quantitative analysis technique using ¹⁸F-flutemetamol-PET calculated SUVr and CL automatically without anatomical images. Moreover, it objectively and homogeneously interpreted positive or negative Aβ burden in the brain as a supplemental tool for the visual reading of equivocal cases in routine clinical practice.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12149-022-01769-x.

60 Years of Achievements by KSNM in Neuroimaging Research

Nuclear medicine neuroimaging is able to show functional and molecular biologic abnormalities in various neuropsychiatric diseases. Therefore, it has played important roles in the clinical diagnosis and in research on the normal and pathological states of the brain. More than 400 outstanding studies have been conducted by Korean researchers over the past 60 years. In the 1990s, when multiheaded single-photon emission computed tomography (SPECT) scanners were first introduced in South Korea, stroke research using brain perfusion SPECT was conducted. With the spread of positron emission tomography (PET) scanners in the 2000s, research on the clinical usefulness of PET and the evaluation of pathophysiology in various diseases such as epilepsy, brain tumors, degenerative brain diseases, and other neuropsychiatric diseases were actively conducted using [¹⁸F]FDG and various neuroreceptor tracers. In the 2010s, with the clinical application of new radiopharmaceuticals for amyloid and tau imaging, research demonstrating the clinical usefulness of PET imaging and the pathophysiology of dementia has increased rapidly. It is expected that the role of nuclear medicine will expand with the development of new radiopharmaceuticals and analysis technologies, along with the application of artificial intelligence for early and differential diagnosis, and the development of therapeutic agents for degenerative brain diseases.

Deep learning improves utility of tau PET in the study of Alzheimer's disease

INTRODUCTION

Positron emission tomography (PET) imaging targeting neurofibrillary tau tangles is increasingly used in the study of Alzheimer's disease (AD), but its utility may be limited by conventional quantitative or qualitative evaluation techniques in earlier disease states. Convolutional neural networks (CNNs) are effective in learning spatial patterns for image classification.

METHODS

18F-MK6240 (n = 320) and AV-1451 (n = 446) PET images were pooled from multiple studies. We performed iterations with differing permutations of radioligands, heuristics, and architectures. Performance was compared to a standard region of interest (ROI)-based approach on prediction of memory impairment. We visualized attention of the network to illustrate decision making.

RESULTS

Overall, models had high accuracy (> 80%) with good average sensitivity and specificity (75% and 82%, respectively), and had comparable or higher accuracy to the ROI standard. Visualizations of model attention highlight known characteristics of tau radioligand binding.

DISCUSSION

CNNs could improve tau PET's role in early disease and extend the utility of tau PET across generations of radioligands.

A Brief History of Nuclear Medicine Physics, Instrumentation, and Data Sciences in Korea

We review the history of nuclear medicine physics, instrumentation, and data sciences in Korea to commemorate the 60^th anniversary of the Korean Society of Nuclear Medicine. In the 1970s and 1980s, the development of SPECT, nuclear stethoscope, and bone densitometry systems, as well as kidney and cardiac image analysis technology, marked the beginning of nuclear medicine physics and engineering in Korea. With the introduction of PET and cyclotron in Korea in 1994, nuclear medicine imaging research was further activated. With the support of large-scale government projects, the development of gamma camera, SPECT, and PET systems was carried out. Exploiting the use of PET scanners in conjunction with cyclotrons, extensive studies on myocardial blood flow quantification and brain image analysis were also actively pursued. In 2005, Korea's first domestic cyclotron succeeded in producing radioactive isotopes, and the cyclotron was provided to six universities and university hospitals, thereby facilitating the nationwide supply of PET radiopharmaceuticals. Since the late 2000s, research on PET/MRI has been actively conducted, and the advanced research results of Korean scientists in the fields of silicon photomultiplier PET and simultaneous PET/MRI have attracted significant attention from the academic community. Currently, Korean researchers are actively involved in endeavors to solve a variety of complex problems in nuclear medicine using artificial intelligence and deep learning technologies.

Artificial Intelligence in Chemistry: Current Trends and Future Directions

The application of artificial intelligence (AI) to chemistry has grown tremendously in recent years. In this Review, we studied the growth and distribution of AI-related chemistry publications in the last two decades using the CAS Content Collection. The volume of both journal and patent publications have increased dramatically, especially since 2015. Study of the distribution of publications over various chemistry research areas revealed that analytical chemistry and biochemistry are integrating AI to the greatest extent and with the highest growth rates. We also investigated trends in interdisciplinary research and identified frequently occurring combinations of research areas in publications. Furthermore, topic analyses were conducted for journal and patent publications to illustrate emerging associations of AI with certain chemistry research topics. Notable publications in various chemistry disciplines were then evaluated and presented to highlight emerging use cases. Finally, the occurrence of different classes of substances and their roles in AI-related chemistry research were quantified, further detailing the popularity of AI adoption in the life sciences and analytical chemistry. In summary, this Review offers a broad overview of how AI has progressed in various fields of chemistry and aims to provide an understanding of its future directions.

Influence of Physical Activity Levels and Functional Capacity on Brain β-Amyloid Deposition in Older Women

Physical activity (PA) and Alzheimer's disease are associated. However, how PA influences the cerebral β-amyloid (Aβ) burden remains unclear. The aim of this study was to determine if PA levels and/or functional capacity (FC) are associated with Aβ plaque deposition, and whether these associations differed according to APOE-ε4 genotype. A total of 117 women (69.7 ± 2.6 years; 33.3% APOE-ε4-carriers) from the Women's Healthy Ageing Project cohort (WHAP) were analyzed. PA was measured using the International Physical Activity Questionnaire and, FC was evaluated using the Timed Up and Go test (TUGt). Positron emission tomography with F-18 Florbetaben was carried out to assess cerebral Aβ burden, and quantified using standardized uptake value rations. The sample was split into PA and TUGt tertiles (T1, T2 and T3), and compared according to APOE-ε4 genotype (positive/negative). There were no significant differences in Aβ accumulation according to PA tertiles and APOE-ε4 genotype. Regarding FC, APOE-ε4+ participants in the first TUGt tertile (high performance) obtained significant lower Aβ accumulations compared with the other two tertiles (p < 0.05). Comparing between genotypes, greater Aβ depositions were found between T2 and T3 in APOE-ε4+ compared with those who were APOE-ε4– (p < 0.05). Values of TUGt ≥ 6.5 s (APOE-ε4+) and 8.5 s (APOE-ε4–) were associated with an increased risk of having higher Aβ retention. In conclusion, low performance in TUGt is associated with a negative effect on brain pathology with increasing cerebral Aβ depositions in older women who are APOE-ε4+. In physically active older women (> 600 METs·min/week), higher PA levels are not associated with reduction in Aβ depositions.