In this paper, we propose a bi-modality medical image synthesis approach based on sequential generative adversarial network (GAN) and semi-supervised learning. Our approach consists of two generative modules that synthesize images of the two modalities in a sequential order. A method for measuring the synthesis complexity is proposed to automatically determine the synthesis order in our sequential GAN. Images of the modality with a lower complexity are synthesized first, and the counterparts with a higher complexity are generated later. Our sequential GAN is trained end-to-end in a semi-supervised manner. In supervised training, the joint distribution of bi-modality images are learned from real paired images of the two modalities by explicitly minimizing the reconstruction losses between the real and synthetic images. To avoid overfitting limited training images, in unsupervised training the marginal distribution of each modality is learned based on unpaired images by minimizing the Wasserstein distance between the distributions of real and fake images. We comprehensively evaluate the proposed model using two synthesis tasks based on three types of evaluate metrics and user studies. Visual and quantitative results demonstrate the superiority of our method to the state-of-the-art methods, and reasonable visual quality and clinical significance. Code is made publicly available at https://github.com/hustlinyi/Multimodal-Medical-Image-Synthesis.

This study aimed to investigate the relationship between pancreatic fat infiltration (PFI) and glucose metabolism disorder, β-cell function and insulin resistance in patients with obesity. Pancreatic fat fraction (PFF) was quantified by MRI IDEAL-IQ technique. PFF greater than 6.2 % was defined as PFI, and 34 obese patients were divided into PFI and non-PFI groups. The 5-point plasma glucose and insulin values during oral glucose tolerance test (OGTT) were recorded. OGTT-derived indices of insulin resistance and β-cell function were calculated. Glucose values levels at 0-120 min during OGTT were significantly higher and β-cell function variables were lower in PFI group than non-PFI group. While indices of insulin resistance were not significantly different between two groups. Correlation analysis showed that PFF was positively correlated with glucose levels at 0, 30 and 60 min, negatively correlated with β-cell function variables and not significantly correlated with indices of insulin resistance. However, these associations of PFF with β-cell function and glucose levels were only present in type 2 diabetes mellitus (T2DM) group but not in non-T2DM group. There is an association between PFI and impaired β-cell function, and increased pancreatic fat may be a potential risk factor for the development of T2DM.

Purpose: To investigate the prognostic value of platelet-to-lymphocyte ratio (PLR) in patients with unresectable hepatocellular carcinoma (uHCC) treated with transarterial chemoembolization (TACE) and tailored tyrosine kinase inhibitors (TKIs) plus immune checkpoints inhibitors (ICIs).Materials and methods: Ninety-eight patients from May 2018 to January 2022 in our hospital were enrolled in this study. The receiver operating characteristic (ROC) curve analysis was performed and the corresponding Youden index was used to determine the optimal PLR cut-off. Overall survival (OS), progression-free survival (PFS), and adverse events (AEs) of patients were evaluated based on the PLR cut-off. The factors affecting survival were assessed using univariate and multivariate Cox proportional hazards regression analyses.Results: The PLR cut-off was 98.89. There were 49 patients in the low pretreatment PLR group (PLR ≤ 98.89) and 49 patients in the high PLR group (PLR > 98.89). Patients with low pretreatment PLR had significantly longer median OS (25.7 months vs 16.1 months; P < 0.001) and PFS (14.9 months vs 10.2 months; P < 0.001) than those with high pretreatment PLR. The multivariate analysis revealed that ALT, tumor size, and PLR are risk factors affecting OS. The three independent factors affecting PFS are tumor size, AFP, and PLR. The AEs were tolerable and manageable.Conclusion: The low pretreatment PLR (PLR ≤ 98.89) was an independent protective factor for the survival outcomes of patients in this study. PLR was helpful for clinicians to predict the prognosis and identify the patients with uHCC who were most likely to benefit from TACE + TKIs + ICIs.

Vertebral landmark localization is a crucial step in various spine-related clinical applications, which requires detecting the corner points of 17 vertebrae. However, the neighboring landmarks often disturb each other because of the homogeneous appearance of vertebrae, making vertebral landmark localization extremely difficult. In this paper, we propose a multi-stage cascaded convolutional neural network (CNN) to split a single task into two sequential steps: center point localization to roughly locate 17 center points of vertebrae, and corner point localization to determine four corner points for each vertebra without any disturbance. The landmarks in each step were located gradually from a set of initialized points by regressing offsets using cascaded CNNs. To resist the mutual attraction of the vertebrae, principal component analysis was employed to preserve the shape constraint in offset regression. We evaluated our method on the AASCE dataset, comprising 609 tight spinal anteroposterior X-ray images, and each image contained 17 vertebrae composed of the thoracic and lumbar spine for spinal shape characterization. The experimental results demonstrated the superior performance of vertebral landmark localization over other state-of-the-art methods, with the relative error decreasing from $3.2e-3$ to $7.2e-4$.