A deep learning model based on contrast-enhanced computed tomography for differential diagnosis of gallbladder carcinoma.

BACKGROUND: Gallbladder carcinoma (GBC) is highly malignant, and its early diagnosis remains difficult. This study aimed to develop a deep learning model based on contrast-enhanced computed tomography (CT) images to assist radiologists in identifying GBC. METHODS: We retrospectively enrolled 278 patients with gallbladder lesions (> 10 mm) who underwent contrast-enhanced CT and cholecystectomy and divided them into the training (n = 194) and validation (n = 84) datasets. The deep learning model was developed based on ResNet50 network. Radiomics and clinical models were built based on support vector machine (SVM) method. We comprehensively compared the performance of deep learning, radiomics, clinical models, and three radiologists. RESULTS: Three radiomics features including LoG_3.0 gray-level size zone matrix zone variance, HHL first-order kurtosis, and LHL gray-level co-occurrence matrix dependence variance were significantly different between benign gallbladder lesions and GBC, and were selected for developing radiomics model. Multivariate regression analysis revealed that age ≥ 65 years [odds ratios (OR) = 4.4, 95% confidence interval (CI): 2.1-9.1, P < 0.001], lesion size (OR = 2.6, 95% CI: 1.6-4.1, P < 0.001), and CA-19-9 > 37 U/mL (OR = 4.0, 95% CI: 1.6-10.0, P = 0.003) were significant clinical risk factors of GBC. The deep learning model achieved the area under the receiver operating characteristic curve (AUC) values of 0.864 (95% CI: 0.814-0.915) and 0.857 (95% CI: 0.773-0.942) in the training and validation datasets, which were comparable with radiomics, clinical models and three radiologists. The sensitivity of deep learning model was the highest both in the training [90% (95% CI: 82%-96%)] and validation [85% (95% CI: 68%-95%)] datasets. CONCLUSIONS: The deep learning model may be a useful tool for radiologists to distinguish between GBC and benign gallbladder lesions.

Physiological and developmental traits associated with the grain yield of winter wheat as affected by phosphorus fertilizer management.

Although researchers have determined that attaining high grain yields of winter wheat depends on the spike number and the shoot biomass, a quantitative understanding of how phosphorus (P) nutrition affects spike formation, leaf expansion and photosynthesis is still lacking. A 3-year field experiment with wheat with six P application rates (0, 25, 50, 100, 200, and 400 kg P ha-1) was conducted to investigate this issue. Stem development and mortality, photosynthetic parameters, dry matter accumulation, and P concentration in whole shoots and in single tillers were studied at key growth stages for this purpose. The results indicated that spike number contributed the most to grain yield of all the yield components in a high-yielding (>8 t/ha) winter wheat system. The main stem (MS) contributed 79% to the spike number and tiller 1 (T1) contributed 21%. The 2.7 g kg-1 tiller P concentration associated with 15 mg kg-1 soil Olsen-P at anthesis stage led to the maximal rate of productive T1s (64%). The critical shoot P concentration that resulted in an adequate product of Pn and LAI was identified as 2.1 g kg-1. The thresholds of shoot P concentration that led to the maximum productive ability of T1 and optimal canopy photosynthetic capacity at anthesis were very similar. In conclusion, the thresholds of soil available P and shoot P concentration in whole plants and in single organs (individual tillers) were established for optimal spike formation, canopy photosynthetic capacity, and dry matter accumulation. These thresholds could be useful in achieving high grain yields while avoiding excessive P fertilization.

Targeting carbon nanotubes based on IGF-1R for photothermal therapy of orthotopic pancreatic cancer guided by optical imaging.

Pancreatic cancer is one of the most lethal malignancies worldwide. The existing therapeutic regimen in the clinic for advanced inoperable carcinomas are far from satisfactory, thus it is urgent to seek more effective anticancer strategies. In the pursuit of novel, more effective interventions, photothermal therapy (PTT) based on nanomaterials has attracted increased attention. Recent advances in related fields have catalyzed the generation of novel nanoprobes, such as organic dyes, metal nanoparticles. However, organic dyes are poorly stable and easy to quench while metal nanoparticles with potential metal toxicity are difficult to degrade, both of which have low light-to-heat conversion efficiency, broad spectrum of anti-tumor effects, and lack of tumor targeting specificity. Single-walled carbon nanotubes (SWNTs) can remedy the above inadequacies. Herein, we report our water-soluble, bio-stable and low-toxicity SWNTs with excellent photothermal conversion efficiency. Specific modifications can enable visualization of the aggregate characteristics of SWNTs at the macroscopic or microscopic level in tumors. The dye-conjugated SWNTs bound with targeting antibodies that can induce them specifically targeting to pancreatic tumors for purposes of performing dyes imaging-guided cytotoxic PTT. PTT using this method achieves precise and excellent curative effects with minimal adverse effects, thus providing a promising strategy for anticancer therapy.