Deep learning-extracted CT imaging phenotypes predict response to total resection in colorectal cancer.

BACKGROUND: Deep learning surpasses many traditional methods for many vision tasks, allowing the transformation of hierarchical features into more abstract, high-level features. PURPOSE: To evaluate the prognostic value of preoperative computed tomography (CT) image texture features and deep learning self-learning high-throughput features (SHF) on postoperative overall survival in the treatment of patients with colorectal cancer (CRC). MATERIAL AND METHODS: The dataset consisted of 810 enrolled patients with CRC confirmed from 10 November 2011 to 10 February 2018. In contrast, SHF extracted by deep learning with multi-task training mechanism and texture features were extracted from the CT with tumor volume region of interest, respectively, and combined with the Cox proportional hazard (CoxPH) model for initial validation to obtain a RAD score to classify patients into high- and low-risk groups. The SHF stability was further validated in combination with Neural Multi-Task Logistic Regression (N-MTLR) model. The overall recognition ability and accuracy of CoxPH and N-MTLR model were evaluated by C-index and Integrated Brier Score (IBS). RESULTS: SHF had a more significant degree of differentiation than texture features. The result is (SHF vs. texture features: C-index: 0.884 vs. 0.611; IBS: 0.025 vs. 0.073) in the CoxPH model, and (SHF vs. texture features: C-index: 0.861 vs. 0.630; IBS: 0.024 vs. 0.065) in N-MTLR. CONCLUSION: SHF is superior to texture features and has potential application for the preoperative prediction of the individualized treatment of CRC.

[Indirect Photodegradation of Sulfamethoxazole in Water].

Indirect photodegradation is one of the primary approaches for the removal of pharmaceutical contaminants from water. This degradation process is dominated by chromophoric dissolved organic matter (CDOM). After illumination, CDOM produces many reactive intermediates, which can react with drug pollutants to achieve indirect photodegradation. In this article, we focused on four different sources of CDOM and factors affecting indirect sulfamethoxazole (SMZ) photolysis. The results show that the indirect photodegradation effect of SMZ is significantly influenced by CDOM. This indirect photodegradation has a dual nature. It promotes the indirect photodegradation of SMZ through the formation of various reactive intermediates and at the same time inhibits the photodegradation of SMZ through light shielding and masking of reactive intermediates. The indirect photodegradation of SMZ is mainly controlled by active intermediates such as 3CDOM*, HO·, and 1O2 produced by CDOM; 3CDOM* is the main participant in indirect photodegradation of SMZ. In addition, the pH, salinity, and nitrate ions have a significant effect on the indirect photodegradation of SMZ, while bicarbonate ions have no significant effect on the indirect photodegradation of SMZ.

Aneuploidy analysis of non-pronuclear embryos from IVF with use of array CGH: a case report.

By using array comparative genomic hybridization (array CGH), to analyze the aneuploidy of the single blastomeres from non-pronuclear embryos on cleavage-stage in IVF cycle. Four non-pronuclear embryos were got from an IVF cycle, and the each single cell was biopsied from the four cleavage-stage embryos on the third day after the insemination which was investigated by using array CGH. After the biopsy, all the embryos continued to cleave, and lately entered the morula stage on the fifth day, just one embryo 3 was developed to early blastocyst stage on the sixth day. The four blastomere 24 chromosomes showed one X monomer and three normal XY diploids; the autosome chromosomes of blastomeres were abnormally gained or lost at different chromosome from four embryos, such as Embryo 1 : 49,X (-1, -5, -11, -19, -20, -21, -Y, +3, +6, +7, +8, +10, +13, +14, +16, +17, +18); Embryo 2 : 44,XY (-12, -15); Embryo 3: 47,XY (-3, -8, -9, -21, +7, +17, +18, +19, +20); Embryo 4 : 54,XY (+4, +7, +10, +12, +13, +16, +17, +22). With the use of the array CGH, the aneuploidy analysis could review the abnormal chromosomes of single blastomere from the non-pronuclear embryos, which can harbor the risk of abnormal sex chromosome and autosome chromosomes.

Magnetic resonance imaging features of vulnerable plaques in an atherosclerotic rabbit model.

BACKGROUND: Noninvasive detection of vulnerable plaque has a significant implication for prevention and treatment of atherosclerotic diseases. The aim of this study is to investigate the difference between vulnerable plaques and stable plaques in magnetic resonance (MR) images. METHODS: Atherosclerosis was induced in twenty male New Zealand white rabbits by high cholesterol diet and balloon injury of the abdominal aorta. After baseline (pre-triggering) MR imaging (MRI) scan, the rabbits underwent pharmaceutical triggering with Russell's viper venom and histamine to induce atherothrombosis, followed by another MRI scan 48 hours later (post-triggering). Rabbits were euthanized to obtain pathological and histological data. The results of MRI were compared with those of pathology and histology. RESULTS: MRI showed that abdominal aorta of the rabbits had pathological change of atherosclerosis in different degrees. Seventy-five plaques were analysed, among which 14 had vulnerable thrombi and 61 stable. Thrombosis was identified in 7 of 11 rabbits by post-triggering MRI, the sensitivity and K value of MR in detection of vulnerable plaque was 71% and 0.803 (P < 0.05). MRI data significantly correlated with the histopathological data in fibrous cap thickness (r = 0.749) plaque area (r = 0.853), lipid core area (r = 0.900). Compared with stable plaques, vulnerable plaques had a significantly thinner fibrous cap ((0.58 ± 0.27) mm vs. (0.95 ± 0.22) mm), larger lipid core area ((7.56 ± 2.78) mm(2) vs. (3.29 ± 1.75) mm(2)), and a higher ratio of lipid core area/plaque area ((55 ± 16)% vs. (27 ± 17)%), but plaque area was comparable in two groups on MRI. The ratio of lipid core area/plaque area was a strong predictor of vulnerable plaques. CONCLUSION: MRI could distinguish vulnerable plaques from stable plaques in a rabbit model of atherothrombosis and may thus be useful as a noninvasive modality for detection of vulnerable plaques in humans.