Clinical evaluation of deep learning-enhanced lymphoma pet imaging with accelerated acquisition.

PURPOSE: This study aims to evaluate the clinical performance of a deep learning (DL)-enhanced two-fold accelerated PET imaging method in patients with lymphoma. METHODS: A total of 123 cases devoid of lymphoma underwent whole-body 18F-FDG-PET/CT scans to facilitate the development of an advanced SAU2Net model, which combines the advantages of U2Net and attention mechanism. This model integrated inputs from simulated 1/2-dose (0.07 mCi/kg) PET acquisition across multiple slices to generate an estimated standard dose (0.14 mCi/kg) PET scan. Additional 39 cases with confirmed lymphoma pathology were utilized to evaluate the model's clinical performance. Assessment criteria encompassed peak-signal-to-noise ratio (PSNR), structural similarity index (SSIM), a 5-point Likert scale rated by two experienced physicians, SUV features, image noise in the liver, and contrast-to-noise ratio (CNR). Diagnostic outcomes, including lesion numbers and Deauville score, were also compared. RESULTS: Images enhanced by the proposed DL method exhibited superior image quality (P < 0.001) in comparison to low-dose acquisition. Moreover, they illustrated equivalent image quality in terms of subjective image analysis and lesion maximum standardized uptake value (SUVmax) as compared to the standard acquisition method. A linear regression model with y = 1.017x + 0.110 ( R 2 = 1.00 ${R^2} = \;1.00$ ) can be established between the enhanced scans and the standard acquisition for lesion SUVmax. With enhancement, increased signal-to-noise ratio (SNR), CNR, and reduced image noise were observed, surpassing those of the standard acquisition. DL-enhanced PET images got diagnostic results essentially equavalent to standard PET images according to two experienced readers. CONCLUSION: The proposed DL method could facilitate a 50% reduction in PET imaging duration for lymphoma patients, while concurrently preserving image quality and diagnostic accuracy.

Fast SPECT/CT planar bone imaging enabled by deep learning enhancement.

BACKGROUND: The application of deep learning methods in rapid bone scintigraphy is increasingly promising for minimizing the duration of SPECT examinations. Recent works showed several deep learning models based on simulated data for the synthesis of high-count bone scintigraphy images from low-count counterparts. Few studies have been conducted and validated on real clinical pairs due to the misalignment inherent in multiple scan procedures. PURPOSE: To generate high quality whole-body bone images from 2× and 3× fast scans using deep learning based enhancement method. MATERIALS AND METHODS: Seventy-six cases who underwent whole-body bone scans were enrolled in this prospective study. All patients went through a standard scan at a speed of 20 cm/min, which followed by fast scans consisting of 2× and 3× accelerations at speeds of 40 and 60 cm/min. A content-attention image restoration approach based on Residual-in-Residual Dense Block (RRDB) is introduced to effectively recover high-quality images from fast scans with fine-details and less noise. Our approach is robust with misalignment introduced from patient's metabolism, and shows valid count-level consistency. Learned Perceptual Image Patch Similarity (LPIPS) and Fréchet Inception Distance (FID) are employed in evaluating the similarity to the standard bone images. To further prove our method practical in clinical settings, image quality of the anonymous images was evaluated by two experienced nuclear physicians on a 5-point Likert scale (5 =  excellent) . RESULTS: The proposed method reaches the state-of-the-art performance on FID and LPIPS with 0.583 and 0.176 for 2× fast scans and 0.583 and 0.185 for 3× fast scans. Clinic evaluation further demonstrated the restored images had a significant improvement compared to fast scan in image quality, technetium 99m-methyl diphosphonate (Tc-99 m MDP) distribution, artifacts, and diagnostic confidence. CONCLUSIONS: Our method was validated for accelerating whole-body bone scans by introducing real clinical data. Confirmed by nuclear medicine physicians, the proposed method can effectively enhance image diagnostic value, demonstrating potential for efficient high-quality fast bone imaging in practical settings.

A dual-attention based coupling network for diabetes classification with heterogeneous data.

Diabetes Mellitus (DM) is a group of metabolic disorders characterized by hyperglycaemia in the absence of treatment. Classification of DM is essential as it corresponds to the respective diagnosis and treatment. In this paper, we propose a new coupling network with hierarchical dual-attention that utilizes heterogeneous data, including Flash Glucose Monitoring (FGM) data and biomarkers in electronic medical records. The long short-term memory-based FGM sub-network extracts the time-dependent features of dynamic FGM sequences, while the biomarkers sub-network learns the features of static biomarkers. The convolutional block attention module (CBAM) for dispersing the feature weights of the spatial and channel dimensions is implemented into the FGM sub-network to endure the variability of FGM and allows us to extract high-level discriminative features more accurately. To better adjust the importance weights of the characteristics of the two sub-networks, self-attention is introduced to integrate the characteristics of heterogeneous data. Based on the dataset provided by Peking University People's Hospital, the proposed method is evaluated through factorial experiments of multi-source heterogeneous data, ablation studies of various attention strategies, time consumption evaluation and quantitative evaluation. The benchmark tests reveal the proposed network achieves a type 1 and 2 diabetes classification accuracy of 95.835% and the comprehensive performance metrics, including Matthews correlation coefficient, F1-score and G-mean, are 91.333%, 94.939% and 94.937% respectively. In the factorial experiments, the proposed method reaches the maximum area under the receiver operating characteristic curve of 0.9428, which indicates the effectiveness of the coupling between the nominated sub-networks. The coupling network with a dual-attention strategy performs better than the one without or only with a single-attention strategy in the ablation study as well. In addition, the model is also tested on another data set, and the accuracy of the test reaches 94.286%, reflecting that the model is robust when it is transferred to untrained diabetes data. The experimental results show that the proposed method is feasible in the classification of diabetes types. The code is available at https://github.com/bitDalei/Diabetes-Classification-with-Heterogeneous-Data.