Detection of Alzheimer's disease using ECD SPECT images by transfer learning from FDG PET.

OBJECTIVE: To develop a practical method to rapidly utilize a deep learning model to automatically extract image features based on a small number of SPECT brain perfusion images in general clinics to objectively evaluate Alzheimer's disease (AD). METHODS: For the properties of low cost and convenient access in general clinics, Tc-99-ECD SPECT imaging data in brain perfusion detection was used in this study for AD detection. Two-stage transfer learning based on the Inception v3 network model was performed using the ImageNet dataset and ADNI database. To improve training accuracy, the three-dimensional image was reorganized into three sets of two-dimensional images for data augmentation and ensemble learning. The effect of pre-training parameters for Tc-99m-ECD SPECT image to distinguish AD from normal cognition (NC) was investigated, as well as the effect of the sample size of F-18-FDG PET images used in pre-training. The same model was also fine-tuned for the prediction of the MMSE score from the Tc-99m-ECD SPECT image. RESULTS: The AUC values of w/wo pre-training parameters for Tc-99m-ECD SPECT image to distinguish AD from NC were 0.86 and 0.90. The sensitivity, specificity, precision, accuracy, and F1 score were 100%, 75%, 76%, 86%, and 86%, respectively for the training model with 1000 cases of F-18-FDG PET image for pre-training. The AUC values for various sample sizes of the training dataset (100, 200, 400, 800, 1000 cases) for pre-training were 0.86, 0.91, 0.95, 0.97, and 0.97. Regardless of the pre-training condition ECD dataset used, the AUC value was greater than 0.85. Finally, predicting cognitive scores and MMSE scores correlated (R2 = 0.7072). CONCLUSIONS: With the ADNI pre-trained model, the sensitivity and accuracy of the proposed deep learning model using SPECT ECD perfusion images to differentiate AD from NC were increased by approximately 30% and 10%, respectively. Our study indicated that the model trained on PET FDG metabolic imaging for the same disease could be transferred to a small sample of SPECT cerebral perfusion images. This model will contribute to the practicality of SPECT cerebral perfusion images using deep learning technology to objectively recognize AD.

High serum uric acid concentration predicts poor survival in patients with breast cancer.

BACKGROUND: Uric acid is a product of purine metabolism. Recently, uric acid has gained much attraction in cancer. In this study, we aim to investigate the clinicopathological and prognostic significance of serum uric acid concentration in breast cancer patients. METHODS: A total of 443 female patients with histopathologically diagnosed breast cancer were included. After a mean follow-up time of 56months, survival was analysed using the Kaplan-Meier method. To further evaluate the prognostic significance of uric acid concentrations, univariate and multivariate Cox regression analyses were applied. RESULTS: Of the clinicopathological parameters, uric acid concentration was associated with age, body mass index, ER status and PR status. Univariate analysis identified that patients with increased uric acid concentration had a significantly inferior overall survival (HR 2.13, 95% CI 1.15-3.94, p=0.016). In multivariate analysis, we found that high uric acid concentration is an independent prognostic factor predicting death, but insufficient to predict local relapse or distant metastasis. Kaplan-Meier analysis indicated that high uric acid concentration is related to the poor overall survival (p=0.013). CONCLUSIONS: High uric acid concentration predicts poor survival in patients with breast cancer, and might serve as a potential marker for appropriate management of breast cancer patients.

Culturing adult human bone marrow stem cells on gelatin scaffold with pNIPAAm as transplanted grafts for skin regeneration.

Skin tissue engineering is a possible solution for the treatment of extensive skin defect. The ultimate goal of skin tissue engineering is to restore the complete functions of native skin, but until now the structures and functions of skins are only partially restored. By negative immunoselection (CD45 and glycophorin A), we isolated and cultivated adult human bone marrow stem cells (hBMSCs) that are of multilineage differentiation potential. In this study, we first demonstrated that by using gelatin/thermo-sensitive poly N-isopropylacrylamide (pNIPAAm) and the immunocompromised mice model, the hBMSCs possess the differentiation potential of epidermis and the capability of healing skin wounds. The in vitro observations and the results of the scanning electron microscope showed that the hBMSCs can attach and proliferate in the gelatin/thermo-sensitive pNIPAAm. To further monitor the in vivo growth effect of the hBMSCs in the skin-defected nude mice, the green fluorescence protein (GFP) gene was transduced into the hBMSCs by the murine stem cell viral vector. The results showed that the rates of cell growth and wound recovery in the hBMSC-treated group were significantly higher than those in the control group, which was only treated with the gelatin/pNIPAAm (p < 0.01). More importantly, the re-epithelialization markers of human pan-cytokeratin and E-cadherin were significantly increased on day 7, day 14, and day 21 after the hBMSC-scaffold with the pNIPAAM in the mice with skin defects (p < 0.05). Moreover, the stem cell markers of human CD13 and CD105 were gradually decreased during the period of wound healing. In sum, this novel method provides a transferring system for cell therapies and maintains its temperature-sensitive property of easy-peeling by lower-temperature treatment. In addition, the in vitro and in vivo GFP imaging systems provide a new imaging modality for understanding the differentiation process and the effective expression of stem cells in wound healing.

Combretastatin A4-induced differential cytotoxicity and reduced metastatic ability by inhibition of AKT function in human gastric cancer cells.

Combretastatin A4 (CA4) is a drug that targets tumor vasculature to inhibit angiogenesis. Whether CA4 has a direct effect on gastric cancer is not known. We herein investigated the effect of CA4 on growth and metastasis of gastric cancer cells at clinically achievable concentration and explored the associated antitumor mechanisms. Nine human gastric cancer cell lines, including two metastatic gastric cancer cell lines (AGS-GFPM1/2), constitutively expressing green fluorescence protein (GFP) were used. These metastatic AGS-GFPM1/2 cells expressed a higher level of phosphorylated serine 473 on AKT (p-AKT). Our results showed that CA4 (0.02-20 microM) has significant in vitro effects on reducing cell attachment, migration, invasiveness, as well as cell cycle G2/M disturbance on p-AKT-positive gastric cancer cells. In addition, a phosphoinositide 3-kinase inhibitor, LY294002 [2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride], a specific AKT inhibitor, and 0.2 to 20 microM CA4 displayed a similar response profile on p-AKT-positive cells, suggesting that CA4-induced effect was mediated by inhibition of the PI3 kinase/AKT pathway. The results from in vivo GFP monitoring system indicated that CA4 phosphate (CA4-P; 200 mg/kg) significantly inhibited the s.c. and intra-abdominal growth of xenotransplanted AGS-GFPM2 cells in nude mice. Furthermore, CA4-P treatment showed a remarkable ability to inhibit gastric tumor metastasis as well as attenuate p-AKT expression. In conclusion, our study is the first to find that CA4 inhibited AKT activity in human gastric cancer cells. The decreased AKT activity correlated well with the CA4 antitumor growth response and decrease of metastasis. Further investigation on drugs targeting the PI3 kinase-AKT pathway may provide a new approach for the treatment of human gastric cancer.