Prediction of 30-day unplanned hospital readmission through survival analysis.

Background and Objective: Unplanned hospital readmissions are a severe and recurrent problem that affects all health systems. Estimating the risk of being readmitted the following days after discharge is difficult since many heterogeneous factors can influence this. The extensive work concerning this problem proposes solutions mostly based on classification machine-learning models. Survival analysis methods could make a better match with the assessment of readmission risk and are yet to become well-established in this field. Methods: We compare different statistical and machine learning survival analysis models trained with right-censored all-cause hospital admission data with covariates available at the moment of discharge. The main focus is on tree-ensemble regression methods based on the assumption of proportional hazards. These models are more thoroughly evaluated at a 30-day time period after discharge, although the actual prediction could be set to any time up to 90 days. Results: The mean performance obtained by each of the proposed survival models ranges from 0.707 to 0.716 C-Index and 0.709 to 0.72 ROC-AUC at a 30-day time period after discharge. The model with the lower performance on both metrics was Cox Proportional Hazards, while the model marking the upper end on both ranges is an XGBoost Regression model with a Cox objective function. Conclusions: Our findings indicate that survival models perform well addressing the hospital readmission problem, machine-learning models getting the edge over statistical methods. There seems to be an improvement over classification models when attempting to predict at a 30-day period since discharge, perhaps due to a better handling of cases nearing the 30-day boundary. Some preprocessing steps, such as limiting the observation period to 90 days after discharge, are also highlighted since they resulted in a performance boost.

SEQENS: An ensemble method for relevant gene identification in microarray data.

This paper describes an ensemble feature identification algorithm called SEQENS, and measures its capability to identify the relevant variables in a case-control study using a genetic expression microarray dataset. SEQENS uses Sequential Feature Search on multiple sample splitting to select variables showing stronger relation with the target, and a variable relevance ranking is finally produced. Although designed for feature identification, SEQENS could also serve as a basis for feature selection (classifier optimisation). Cliff, a ranking evaluation metric is also presented and used to assess the feature identification algorithms when a groundtruth of relevant variables is available. To test performance, three types of synthetic groundtruths emulating fictitious diseases are generated from ten randomly chosen variables following different target pattern distributions using the E-MTAB-3732 dataset. Several sample-to-dimensionality ratios ranging from 300 to 3,000 observations and 854 to 54,675 variables are explored. SEQENS is compared with other feature selection or identification state-of-the-art methods. On average, the proposed algorithm identifies better the relevant genes and exhibits a stronger stability. The algorithm is available to the community.

A deep learning system to obtain the optimal parameters for a threshold-based breast and dense tissue segmentation.

BACKGROUND AND OBJECTIVE: Breast cancer is the most frequent cancer in women. The Spanish healthcare network established population-based screening programs in all Autonomous Communities, where mammograms of asymptomatic women are taken with early diagnosis purposes. Breast density assessed from digital mammograms is a biomarker known to be related to a higher risk to develop breast cancer.It is thus crucial to provide a reliable method to measure breast density from mammograms. Furthermore the complete automation of this segmentation process is becoming fundamental as the amount of mammograms increases every day. Important challenges are related with the differences in images from different devices and the lack of an objective gold standard.This paper presents a fully automated framework based on deep learning to estimate the breast density. The framework covers breast detection, pectoral muscle exclusion, and fibroglandular tissue segmentation. METHODS: A multi-center study, composed of 1785 women whose "for presentation" mammograms were segmented by two experienced radiologists. A total of 4992 of the 6680 mammograms were used as training corpus and the remaining (1688) formed the test corpus. This paper presents a histogram normalization step that smoothed the difference between acquisition, a regression architecture that learned segmentation parameters as intrinsic image features and a loss function based on the DICE score. RESULTS: The results obtained indicate that the level of concordance (DICE score) reached by the two radiologists (0.77) was also achieved by the automated framework when it was compared to the closest breast segmentation from the radiologists. For the acquired with the highest quality device, the DICE score per acquisition device reached 0.84, while the concordance between radiologists was 0.76. CONCLUSIONS: An automatic breast density estimator based on deep learning exhibits similar performance when compared with two experienced radiologists. It suggests that this system could be used to support radiologists to ease its work.

Global parenchymal texture features based on histograms of oriented gradients improve cancer development risk estimation from healthy breasts.

BACKGROUND: The breast dense tissue percentage on digital mammograms is one of the most commonly used markers for breast cancer risk estimation. Geometric features of dense tissue over the breast and the presence of texture structures contained in sliding windows that scan the mammograms may improve the predictive ability when combined with the breast dense tissue percentage. METHODS: A case/control study nested within a screening program covering 1563 women with craniocaudal and mediolateral-oblique mammograms (755 controls and the contralateral breast mammograms at the closest screening visit before cancer diagnostic for 808 cases) aging 45 to 70 from Comunitat Valenciana (Spain) was used to extract geometric and texture features. The dense tissue segmentation was performed using DMScan and validated by two experienced radiologists. A model based on Random Forests was trained several times varying the set of variables. A training dataset of 1172 patients was evaluated with a 10-stratified-fold cross-validation scheme. The area under the Receiver Operating Characteristic curve (AUC) was the metric for the predictive ability. The results were assessed by only considering the output after applying the model to the test set, which was composed of the remaining 391 patients. RESULTS: The AUC score obtained by the dense tissue percentage (0.55) was compared to a machine learning-based classifier results. The classifier, apart from the percentage of dense tissue of both views, firstly included global geometric features such as the distance of dense tissue to the pectoral muscle, dense tissue eccentricity or the dense tissue perimeter, obtaining an accuracy of 0.56. By the inclusion of a global feature based on local histograms of oriented gradients, the accuracy of the classifier was significantly improved (0.61). The number of well-classified patients was improved up to 236 when it was 208. CONCLUSION: Relative geometric features of dense tissue over the breast and histograms of standardized local texture features based on sliding windows scanning the whole breast improve risk prediction beyond the dense tissue percentage adjusted by geometrical variables. Other classifiers could improve the results obtained by the conventional Random Forests used in this study.