Clinical baseline matching is a critical step in the process of transforming real-world clinical records into novel medical knowledge, more precise diagnosis and efficacious treatments. Conventionally, the baseline matching was performed by heuristic methods or even manually. Here, a baseline-matching algorithm called "dynamite" was proposed using the dynamic programming technique. This algorithm minimizes the discrepancy of values of clinical variables or propensity scores between two patient groups in the study, while at the same time includes as many patients as possible so as to maximize statistical power. This algorithm was applied to an anti-hepatitis B research where two classes of approved drugs, peginterferon and nucleos(t)ide reverse transcription inhibitors (NRTI), were compared in terms of their protective effect in patients against hepatocellular carcinoma (HCC). Patients treated by the two classes of medications were retrospectively recruited from clinical records. Initially, age and HBeAg positivity were significantly different between candidate study subjects in the two treatment groups, with 153 and 177 patients respectively. Using the baseline-matching algorithm, the baseline characteristics of the two included patient groups, each comprised 120 patients, were well matched. Longitudinal analysis showed that the peginterferon-treated group achieved better HCC-free survival than the NRTI-treated groups (P = 0.0087293).

This paper addresses two challenging issues in unsupervised multiscale texture segmentation: determining adequate spatial and feature resolutions for different regions of the image, and utilizing information across different scales/resolutions. The center of a homogeneous texture is analyzed using coarse spatial resolution, and its border is detected using fine spatial resolution so as to locate the boundary accurately. The extraction of texture features is achieved via a multiresolution pyramid. The feature values are integrated across scales/resolutions adaptively. The number of textures is determined automatically using the variance ratio criterion. Experimental results on synthetic and real images demonstrate the improvement in performance of the proposed multiscale scheme over single scale approaches.

We propose a new method that combines the use of a global threshold and a window-based scheme for computing local thresholds. The latter scheme compares the contrast of gray values within a neighborhood whose size varies with the scale of the objects being examined. To compute the scale quantity, a new wavelet model entitled Hadamard multiresolution analysis is also proposed. When the window-based scheme is applied to the areas where the global threshold is likely to fail, we obtain uniformly better binary results than using a global threshold only. Significant improvements to OCR performance can also be achieved by our binary results.

This paper shows that in edge detection the regularization factor alpha is a better scale parameter than the standard deviation (sigma) of the Gaussian pre-filter. The alpha scale space, which exhibits the evolutionary behaviour of an edge in various scales, is the basis for the design of a multiscale edge detector (MRCBS). In MRCBS, the scale is determined adaptively according to the local noise level; the thresholds which control the amount of edge details are adjusted according to the scale; and the anisotropic diffusion is applied in the finest scale to further suppress noise.