Machine learning-based diagnosis for disseminated intravascular coagulation (DIC): Development, external validation, and comparison to scoring systems.

The major challenge in the diagnosis of disseminated intravascular coagulation (DIC) comes from the lack of specific biomarkers, leading to developing composite scoring systems. DIC scores are simple and rapidly applicable. However, optimal fibrin-related markers and their cut-off values remain to be defined, requiring optimization for use. The aim of this study is to optimize the use of DIC-related parameters through machine learning (ML)-approach. Further, we evaluated whether this approach could provide a diagnostic value in DIC diagnosis. For this, 46 DIC-related parameters were investigated for both clinical findings and laboratory results. We retrospectively reviewed 656 DIC-suspected cases at an initial order for full DIC profile and labeled their evaluation results (Set 1; DIC, n = 228; non-DIC, n = 428). Several ML algorithms were tested, and an artificial neural network (ANN) model was established via independent training and testing using 32 selected parameters. This model was externally validated from a different hospital with 217 DIC-suspected cases (Set 2; DIC, n = 80; non-DIC, n = 137). The ANN model represented higher AUC values than the three scoring systems in both set 1 (ANN 0.981; ISTH 0.945; JMHW 0.943; and JAAM 0.928) and set 2 (AUC ANN 0.968; ISTH 0.946). Additionally, the relative importance of the 32 parameters was evaluated. Most parameters had contextual importance, however, their importance in ML-approach was different from the traditional scoring system. Our study demonstrates that ML could optimize the use of clinical parameters with robustness for DIC diagnosis. We believe that this approach could play a supportive role in physicians' medical decision by integrated into electrical health record system. Further prospective validation is required to assess the clinical consequence of ML-approach and their clinical benefit.

FLIM-FRET analyzer: open source software for automation of lifetime-based FRET analysis.

BACKGROUND: Despite the broad use of FRET techniques, available methods for analyzing protein-protein interaction are subject to high labor and lack of systematic analysis. We propose an open source software allowing the quantitative analysis of fluorescence lifetime imaging (FLIM) while integrating the steady-state fluorescence intensity information for protein-protein interaction studies. FINDINGS: Our developed open source software is dedicated to fluorescence lifetime imaging microscopy (FLIM) data obtained from Becker & Hickl SPC-830. FLIM-FRET analyzer includes: a user-friendly interface enabling automated intensity-based segmentation into single cells, time-resolved fluorescence data fitting to lifetime value for each segmented objects, batch capability, and data representation with donor lifetime versus acceptor/donor intensity quantification as a measure of protein-protein interactions. CONCLUSIONS: The FLIM-FRET analyzer software is a flexible application for lifetime-based FRET analysis. The application, the C#. NET source code, and detailed documentation are freely available at the following URL: http://FLIM-analyzer.ip-korea.org.

Analysis of image-based phenotypic parameters for high throughput gene perturbation assays.

Although image-based phenotypic assays are considered a powerful tool for siRNA library screening, the reproducibility and biological implications of various image-based assays are not well-characterized in a systematic manner. Here, we compared the resolution of high throughput assays of image-based cell count and typical cell viability measures for cancer samples. It was found that the optimal plating density of cells was important to obtain maximal resolution in both types of assays. In general, cell counting provided better resolution than the cell viability measure in diverse batches of siRNAs. In addition to cell count, diverse image-based measures were simultaneously collected from a single screening and showed good reproducibility in repetitions. They were classified into a few functional categories according to biological process, based on the differential patterns of hit (i.e., siRNAs) prioritization from the same screening data. The presented systematic analyses of image-based parameters provide new insight to a multitude of applications and better biological interpretation of high content cell-based assays.

Numerical evaluation and comparison of Kalantari's zero bounds for complex polynomials.

In this paper, we investigate the performance of zero bounds due to Kalantari and Dehmer by using special classes of polynomials. Our findings are evidenced by numerical as well as analytical results.

Robust dose-response curve estimation applied to high content screening data analysis.

BACKGROUND AND METHOD: Successfully automated sigmoidal curve fitting is highly challenging when applied to large data sets. In this paper, we describe a robust algorithm for fitting sigmoid dose-response curves by estimating four parameters (floor, window, shift, and slope), together with the detection of outliers. We propose two improvements over current methods for curve fitting. The first one is the detection of outliers which is performed during the initialization step with correspondent adjustments of the derivative and error estimation functions. The second aspect is the enhancement of the weighting quality of data points using mean calculation in Tukey's biweight function. RESULTS AND CONCLUSION: Automatic curve fitting of 19,236 dose-response experiments shows that our proposed method outperforms the current fitting methods provided by MATLAB®;'s nlinfit function and GraphPad's Prism software.

The quality of zero bounds for complex polynomials.

In this paper, we evaluate the quality of zero bounds on the moduli of univariate complex polynomials. We select classical and recently developed bounds and evaluate their quality by using several sets of complex polynomials. As the quality of priori bounds has not been investigated thoroughly, our results can be useful to find optimal bounds to locate the zeros of complex polynomials.