Fabio González

High-throughput adaptive sampling for whole-slide histopathology image analysis (HASHI) via convolutional neural networks: Application to invasive breast cancer detection

Precise detection of invasive cancer on whole-slide images (WSI) is a critical first step in digital pathology tasks of diagnosis and grading. Convolutional neural network (CNN) is the most popular representation learning method for computer vision tasks, which have been successfully applied in digital pathology, including tumor and mitosis detection. However, CNNs are typically only tenable with relatively small image sizes (200 × 200 pixels). Only recently, Fully convolutional networks (FCN) are able to deal with larger image sizes (500 × 500 pixels) for semantic segmentation. Hence, the direct application of CNNs to WSI is not computationally feasible because for a WSI, a CNN would require billions or trillions of parameters. To alleviate this issue, this paper presents a novel method, High-throughput Adaptive Sampling for whole-slide Histopathology Image analysis (HASHI), which involves: i) a new efficient adaptive sampling method based on probability gradient and quasi-Monte Carlo sampling, and, ii) a powerful representation learning classifier based on CNNs. We applied HASHI to automated detection of invasive breast cancer on WSI. HASHI was trained and validated using three different data cohorts involving near 500 cases and then independently tested on 195 studies from The Cancer Genome Atlas. The results show that (1) the adaptive sampling method is an effective strategy to deal with WSI without compromising prediction accuracy by obtaining comparative results of a dense sampling (∼6 million of samples in 24 hours) with far fewer samples (∼2,000 samples in 1 minute), and (2) on an independent test dataset, HASHI is effective and robust to data from multiple sites, scanners, and platforms, achieving an average Dice coefficient of 76%.

XML annotations by MF and NS of WSIs from CINJ

XML region-based annotations by MF and NS pathologists of whole-slide images (WSIs) from CINJ institution data cohort.

Continuous assessment in a computer programming course supported by a software tool

This article presents a continuous assessment methodology for a computer programming course supported by an automatic assessment tool, applied to the practical programming exercises performed by the students. The interaction between the students and the assessment tool was studied through quantitative analyses. In particular, the solutions proposed by the students (computer programs) were analyzed using the verdicts given by the automatic assessment tool: correct solutions or incorrect solutions. In the case of incorrect solutions, the types of programming errors were studied. Additionally, interaction was also studied by analyzing the students’ success rate. This rate is the percentage of correct solutions among the total number of attempts (correct and incorrect). Moreover, the relationship between success rate and academic performance was analyzed. Furthermore, this research examines the students’ perceptions toward the assessment tool through interviews. The results of this study help understanding the benefits and perceptions of the students with respect to the use of an automatic assessment tool in a computer programming course.

A Strategy Based on Technological Maps for the Identification of the State-of-the-Art Techniques in Software Development Projects: Virtual Judge Projects as a Case Study

We propose a novel strategy based on technological watch (TW) to identify the state-of-the-art techniques of software development projects. Taking as a starting point the data analysis of the GitHub platform using the VigHub tool, technological maps that describe the development and evolution of software projects, and perspectives of specific technologies are obtained. The proposed strategy was tested in a case study about virtual judge projects for programming. When analyzing the GitHub data, four main technological maps were obtained: programming languages, topic evolution timeline, successful projects, and successful users and organizations. This article shows how the use of the developed strategy supports the identification of the state-of-the-art software techniques. This facilitates the appropriate identification of ideas, source code, and tools that can support and improve the software development, expanding the creation of strategies to support the decision-making process on new development projects and technological inventions.

Glaucoma Diagnosis from Eye Fundus Images Based on Deep Morphometric Feature Estimation

Glaucoma is an ophthalmic disease related to damage in the optic nerve and it is without symptoms in its early stages. Left untreated, it can lead to vision limitation and blindness. Eye fundus images have been widely accepted by medical personnel to examine the morphology and texture of the optic nerve head and the physiologic cup but glaucoma diagnosis is still subjective and without clear consensus among experts. This paper presents a multi-stage deep learning model for glaucoma diagnosis based on a curriculum learning strategy. In curriculum learning, a model is sequentially trained to solve incrementally difficult tasks. Our proposed model includes the following stages: segmentation of the optic disc and physiological cup, prediction of morphometric features from segmentations, and prediction of disease level (healthy, suspicious and glaucoma). The experimental evaluation shows that our proposed method outperforms conventional convolutional deep learning models from the state of the art reported on the RIM-ONE-v1 and DRISHTI-GS1 datasets with an accuracy of 89.4% and an AUC of 0.82 respectively.