Optimizing allocation of colorectal cancer screening hospitals in Shanghai: a geospatial analysis.

Screening programmes are important for early diagnosis and treatment of colorectal cancer (CRC) but they are not equally efficient in all locations. Depending on which hospital people belong to, they often are not willing to follow up even after a positive result, resulting in a lower-than-expected overall detection rate. Improved allocation of health resources would increase the program's efficiency and assist hospital accessibility. A target population exceeding 70,000 people and 18 local hospitals were included in the investigation of an optimization plan based on a locationallocation model. We calculated the hospital service areas and the accessibility for people in communities to CRC-screening hospitals using the Huff Model and the Two-Step Floating Catchment Area (2SFCA) approach. We found that only 28.2% of the residents with initially a positive screening result had chosen followup with colonoscopy and significant geographical differences in spatial accessibility to healthcare services indeed exist. The lowest accessibility was found in the Southeast, including the Zhangjiang, Jichang and Laogang communities with the best accessibility mainly distributed near the city centre of Lujiazui; the latter also had relatively a high level of what is called "ineffective screening" as it represents wasteful resource allocation. It is recommended that Hudong Hospital should be chosen instead of Punan Hospital as the optimization, which can improve the service population of each hospital and the populations served per colonoscope. Based on our results, changes in hospital configuration in colorectal cancer screening programme are needed to achieve adequate population coverage and equitable facility accessibility. Planning of medical services should be based on the spatial distribution trends of the population served.

Impacts of sample ratio and size on the performance of random forest model to predict the potential distribution of snail habitats.

Few studies have considered the impacts of sample size and sample ratio of presence and absence points on the results of random forest (RF) testing. We applied this technique for the prediction of the spatial distribution of snail habitats based on a total of 15,000 sample points (5,000 presence samples and 10,000 control points). RF models were built using seven different sample ratios (1:1, 1:2, 1:3, 1:4, 2:1, 3:1, and 4:1) and the optimal ratio was identified via the Area Under the Curve (AUC) statistic. The impact of sample size was compared by RF models under the optimal ratio and the optimal sample size. When the sample size was small, the sampling ratios of 1:1, 1:2 and 1:3 were significantly better than the sample ratios of 4:1 and 3:1 at all four levels of sample sizes (p<0.01) and there was no significant difference among the ratios of 1:1, 1:2 and 1:3 (p>0.05). The sample ratio of 1:2 appeared to be optimal for a relatively large sample size with the lowest quartile deviation. In addition, increasing the sample size produced a higher AUC and a smaller slope and the most suitable sample size found in this study was 2400 (AUC=0.96). This study provides a feasible idea to select an appropriate sample size and sample ratio for ecological niche modelling (ENM) and also provides a scientific basis for the selection of samples to accurately identify and predict snail habitat distributions.