AI approach of cycle-consistent generative adversarial networks to synthesize PET images to train computer-aided diagnosis algorithm for dementia

Generative adversarial network-created brain SPECTs of cerebral ischemia are indistinguishable to scans from real patients

Deep convolutional generative adversarial networks (GAN) allow for creating images from existing databases. We applied a modified light-weight GAN (FastGAN) algorithm to cerebral blood flow SPECTs and aimed to evaluate whether this technology can generate created images close to real patients. Investigating three anatomical levels (cerebellum, CER; basal ganglia, BG; cortex, COR), 551 normal (248 CER, 174 BG, 129 COR) and 387 pathological brain SPECTs using N-isopropyl p-I-123-iodoamphetamine (¹²³I-IMP) were included. For the latter scans, cerebral ischemic disease comprised 291 uni- (66 CER, 116 BG, 109 COR) and 96 bilateral defect patterns (44 BG, 52 COR). Our model was trained using a three-compartment anatomical input (dataset 'A'; including CER, BG, and COR), while for dataset 'B', only one anatomical region (COR) was included. Quantitative analyses provided mean counts (MC) and left/right (LR) hemisphere ratios, which were then compared to quantification from real images. For MC, 'B' was significantly different for normal and bilateral defect patterns (P < 0.0001, respectively), but not for unilateral ischemia (P = 0.77). Comparable results were recorded for LR, as normal and ischemia scans were significantly different relative to images acquired from real patients (P ≤ 0.01, respectively). Images provided by 'A', however, revealed comparable quantitative results when compared to real images, including normal (P = 0.8) and pathological scans (unilateral, P = 0.99; bilateral, P = 0.68) for MC. For LR, only uni- (P = 0.03), but not normal or bilateral defect scans (P ≥ 0.08) reached significance relative to images of real patients. With a minimum of only three anatomical compartments serving as stimuli, created cerebral SPECTs are indistinguishable to images from real patients. The applied FastGAN algorithm may allow to provide sufficient scan numbers in various clinical scenarios, e.g., for "data-hungry" deep learning technologies or in the context of orphan diseases.

Applications of Generative Adversarial Networks (GANs) in Positron Emission Tomography (PET) imaging: A review

PURPOSE

This paper reviews recent applications of Generative Adversarial Networks (GANs) in Positron Emission Tomography (PET) imaging. Recent advances in Deep Learning (DL) and GANs catalysed the research of their applications in medical imaging modalities. As a result, several unique GAN topologies have emerged and been assessed in an experimental environment over the last two years.

METHODS

The present work extensively describes GAN architectures and their applications in PET imaging. The identification of relevant publications was performed via approved publication indexing websites and repositories. Web of Science, Scopus, and Google Scholar were the major sources of information.

RESULTS

The research identified a hundred articles that address PET imaging applications such as attenuation correction, de-noising, scatter correction, removal of artefacts, image fusion, high-dose image estimation, super-resolution, segmentation, and cross-modality synthesis. These applications are presented and accompanied by the corresponding research works.

CONCLUSION

GANs are rapidly employed in PET imaging tasks. However, specific limitations must be eliminated to reach their full potential and gain the medical community's trust in everyday clinical practice.

A review on AI in PET imaging

Artificial intelligence (AI) has been applied to various medical imaging tasks, such as computer-aided diagnosis. Specifically, deep learning techniques such as convolutional neural network (CNN) and generative adversarial network (GAN) have been extensively used for medical image generation. Image generation with deep learning has been investigated in studies using positron emission tomography (PET). This article reviews studies that applied deep learning techniques for image generation on PET. We categorized the studies for PET image generation with deep learning into three themes as follows: (1) recovering full PET data from noisy data by denoising with deep learning, (2) PET image reconstruction and attenuation correction with deep learning and (3) PET image translation and synthesis with deep learning. We introduce recent studies based on these three categories. Finally, we mention the limitations of applying deep learning techniques to PET image generation and future prospects for PET image generation.

Risk factors and machine learning model for predicting hospitalization outcomes in geriatric patients with dementia

Introduction

Geriatric patients with dementia incur higher healthcare costs and longer hospital stays than other geriatric patients. We aimed to identify risk factors for hospitalization outcomes that could be mitigated early to improve outcomes and impact overall quality of life.

Methods

We identified risk factors, that is, demographics, hospital complications, pre‐admission, and post‐admission risk factors including medical history and comorbidities, affecting hospitalization outcomes determined by hospital stays and discharge dispositions. Over 150 clinical and demographic factors of 15,678 encounters (8407 patients) were retrieved from our institution's data warehouse. We further narrowed them down to twenty factors through feature selection engineering by using analysis of variance (ANOVA) and Glmnet. We developed an explainable machine‐learning model to predict hospitalization outcomes among geriatric patients with dementia.

Results

Our model is based on stacking ensemble learning and achieved accuracy of 95.6% and area under the curve (AUC) of 0.757. It outperformed prevalent methods of risk assessment for encounters of patients with Alzheimer's disease dementia (ADD) (4993), vascular dementia (VD) (4173), Parkinson's disease with dementia (PDD) (3735), and other unspecified dementias (OUD) (2777). Top identified hospitalization outcome risk factors, mostly from medical history, include encephalopathy, number of medical problems at admission, pressure ulcers, urinary tract infections, falls, admission source, age, race, anemia, etc., with several overlaps in multi‐dementia groups.

Discussion

Our model identified several predictive factors that can be modified or intervened so that efforts can be made to prevent recurrence or mitigate their adverse effects. Knowledge of the modifiable risk factors would help guide early interventions for patients at high risk for poor hospitalization outcome as defined by hospital stays longer than seven days, undesirable discharge disposition, or both. The interventions include starting specific protocols on modifiable risk factors like encephalopathy, falls, and infections, where non‐existent or not routine, to improve hospitalization outcomes of geriatric patients with dementia.

Highlights

A total 15,678 encounters of Geriatrics with dementia with a final 20 risk factors.

Developed a predictive model for hospitalization outcomes for multi‐dementia types.

Risk factors for each type were identified including those amenable to interventions.

Top factors are encephalopathy, pressure ulcers, urinary tract infection (UTI), falls, and admission source.

With accuracy of 95.6%, our ensemble predictive model outperforms other models.

Advancing diagnostic performance and clinical applicability of deep learning-driven generative adversarial networks for Alzheimer's disease

Alzheimer's disease (AD) is a neurodegenerative disease that severely affects the activities of daily living in aged individuals, which typically needs to be diagnosed at an early stage. Generative adversarial networks (GANs) provide a new deep learning method that show good performance in image processing, while it remains to be verified whether a GAN brings benefit in AD diagnosis. The purpose of this research is to systematically review psychoradiological studies on the application of a GAN in the diagnosis of AD from the aspects of classification of AD state and AD-related image processing compared with other methods. In addition, we evaluated the research methodology and provided suggestions from the perspective of clinical application. Compared with other methods, a GAN has higher accuracy in the classification of AD state and better performance in AD-related image processing (e.g. image denoising and segmentation). Most studies used data from public databases but lacked clinical validation, and the process of quantitative assessment and comparison in these studies lacked clinicians' participation, which may have an impact on the improvement of generation effect and generalization ability of the GAN model. The application value of GANs in the classification of AD state and AD-related image processing has been confirmed in reviewed studies. Improvement methods toward better GAN architecture were also discussed in this paper. In sum, the present study demonstrated advancing diagnostic performance and clinical applicability of GAN for AD, and suggested that the future researchers should consider recruiting clinicians to compare the algorithm with clinician manual methods and evaluate the clinical effect of the algorithm.

Generative image transformer (GIT): unsupervised continuous image generative and transformable model for [123I]FP-CIT SPECT images

OBJECTIVE

Recently, generative adversarial networks began to be actively studied in the field of medical imaging. These models are used for augmenting the variation of images to improve the accuracy of computer-aided diagnosis. In this paper, we propose an alternative new image generative model based on transformer decoder blocks and verify the performance of our model in generating SPECT images that have characteristics of Parkinson's disease patients.

METHODS

Firstly, we proposed a new model architecture that is based on a transformer decoder block and is extended to generate slice images. From few superior slices of 3D volume, our model generates the rest of the inferior slices sequentially. Our model was trained by using [¹²³I]FP-CIT SPECT images of Parkinson's disease patients that originated from the Parkinson's Progression Marker Initiative database. Pixel values of SPECT images were normalized by the specific/nonspecific binding ratio (SNBR). After training the model, we generated [¹²³I]FP-CIT SPECT images. The transformation of images of the healthy control case SPECT images into PD-like images was also performed. Generated images were visually inspected and evaluated using the mean absolute value and asymmetric index.

RESULTS

Our model was successfully generated and transformed into PD-like SPECT images. The mean absolute SNBR was mostly less than 0.15 in absolute value. The variation of the obtained dataset images was confirmed by the analysis of the asymmetric index.

CONCLUSIONS

These results showed the potential ability of our new generative approach for SPECT images that the generative model based on the transformer realized both generation and transformation by a single model.

Advanced Machine Learning-Based Analytics on COVID-19 Data Using Generative Adversarial Networks

The domain of medical diagnosis and predictive analytics is one of the key domains of research with enormous dimensions whereby the diseases of different types can be predicted. Nowadays, there is a huge panic of impact and rapid mutation of the COVID-19 virus impression. The world is getting affected by this virus to a huge extent and there is no vaccine developed so far. India is also having more than 10,000 patients with than 300 deceased. The global human community is having around 20 lacs of Coronavirus patients. The Generative Adversarial Network (GAN) is the contemporary high-performance approach in which the use of advanced neural networks is done for the cavernous analytics of the images and multimedia data. In this research work, the analytics of key points from medical images of the COVID-19 dataset is to be presented using which the diagnosis and predictions can be done for the patients. The GANs are used for the generation, transformation as well as presentation of the dataset and key points using advanced deep learning models which can analyze the patterns in the medical images including X-Ray, CT Scan, and many others. Using such approaches with the integration of GANs, the overall predictive analytics can be made high performance aware as compared to the classical neural networks with multiple layers. In this research manuscript, the inscription of work is projected on the benchmark datasets with the advanced scripting so that the predictive mining and knowledge discovery can be done effectively with more accuracy.

Artificial Intelligence and Machine Learning in Nuclear Medicine: Future Perspectives

Artificial intelligence and machine learning based approaches are increasingly finding their way into various areas of nuclear medicine imaging. With the technical development of new methods and the expansion to new fields of application, this trend is likely to become even more pronounced in future. Possible means of application range from automated image reading and classification to correlation with clinical outcomes and to technological applications in image processing and reconstruction. In the context of tumor imaging, that is, predominantly FDG or PSMA PET imaging but also bone scintigraphy, artificial intelligence approaches can be used to quantify the whole-body tumor volume, for the segmentation and classification of pathological foci or to facilitate the diagnosis of micro-metastases. More advanced applications aim at the correlation of image features that are derived by artificial intelligence with clinical endpoints, for example, whole-body tumor volume with overall survival. In nuclear medicine imaging of benign diseases, artificial intelligence methods are predominantly used for automated and/or facilitated image classification and clinical decision making. Automated feature selection, segmentation and classification of myocardial perfusion scintigraphy can help in identifying patients that would benefit from intervention and to forecast clinical prognosis. Automated reporting of neurodegenerative diseases such as Alzheimer's disease might be extended to early diagnosis-being of special interest, if targeted treatment options might become available. Technological approaches include artificial intelligence-based attenuation correction of PET images, image reconstruction or anatomical landmarking. Attenuation correction is of special interest for avoiding the need of a coregistered CT scan, in the process of image reconstruction artefacts might be reduced, or ultra low-dose PET images might be denoised. The development of accurate ultra low-dose PET imaging might broaden the method's applicability, for example, toward oncologic PET screening. Most artificial intelligence approaches in nuclear medicine imaging are still in early stages of development, further improvements are necessary for broad clinical applications. In this review, we describe the current trends in the context fields of body oncology, cardiac imaging, and neuroimaging while an additional section puts emphasis on technological trends. Our aim is not only to describe currently available methods, but also to place a special focus on the description of possible future developments.