Cell recognition based on topological sparse coding for microscopy imaging of focused ultrasound treatment.

BACKGROUND: Ultrasound is considered a reliable, widely available, non-invasive, and inexpensive imaging technique for assessing and detecting the development phases of cancer; both in vivo and ex vivo, and for understanding the effects on cell cycle and viability after ultrasound treatment. METHODS: Based on the topological continuity characteristics, and that adjacent points or areas represent similar features, we propose a topological penalized convex objective function of sparse coding, to recognize similar cell phases. RESULTS: This method introduces new features using a deep learning method of sparse coding with topological continuity characteristics. Large-scale comparison tests demonstrate that the RAW can outperform SIFT GIST and HoG as the input features with this method, achieving higher sensitivity, specificity, F1 score, and accuracy. CONCLUSIONS: Experimental results show that the proposed topological sparse coding technique is valid and effective for extracting new features, and the proposed system was effective for cell recognition of microscopy images of theMDA-MB-231 cell line. This method allows features from sparse coding learning methods to have topological continuity characteristics, and the RAW features are more applicable for the deep learning of the topological sparse coding method than SIFT GIST and HoG.

Cardiorenal syndrome in incident peritoneal dialysis patients: What is its effect on patients' outcomes?

BACKGROUND: Peritoneal dialysis (PD) is increasingly used for long-term management of Cardiorenal Syndrome (CRS). We compared outcomes in incident PD patients according to their baseline heart failure status. METHODS: This retrospective cohort study evaluated all-cause and cardiovascular mortality in incident PD patients with different heart failure status (non-CRS, acute heart failure [AHF], type II CRS, type IV CRS) who started PD between 2006 and 2016 in the Peking University Third Hospital. RESULTS: Of 748 patients included in the study, there were 466 (62.3%), 214 (28.6%), 27 (3.6%), and 41 (5.5%) patients in the non-CRS, AHF, type II CRS and type IV CRS groups, respectively. Patients with CRS were older (p<0.001), with more diabetes mellitus (p<0.001), coronary heart history (p<0.001), higher estimated glomerular filtration rate (eGFR) (p<0.001), lower serum creatinine (p<0.001) and phosphorus levels (p = 0.003) compared to non-CRS patients. Respective all-cause survival rates for patients with non-CRS, AHF, type II CRS and type IV CRS were 90.6%, 87.1%, 85.2% and 84.8% at 1 year, and 63.1%, 47.7%, 27.3% and 35.1% at 5 years (p<0.001). The corresponding figures for cardiovascular survival were 93%, 92%, 84% and 81% at 1 year, and 67%, 59%, 55% and 54% at 5 years (p<0.001). However, after adjusting for confounding factors, the presence of CRS was not independently associated with all-cause mortality whereas type IV CRS (HR 2.10, 95% CI 1.03-4.28, p = 0.04) was associated with higher cardiovascular mortality as compared to without CRS. CONCLUSION: Incident PD patients with different types of CRS had higher rates of both all-cause and cardiovascular mortality compared with patients without CRS. However, these observed adverse outcomes may be related to associated older age and higher prevalence of comorbidities, rather than CRS per se, except for type IV CRS, treatment strategies to reduce high cardiovascular CVD mortality may needed.