Publications

Computer systems have grown in complexity to the point where manual inspection of system behaviour for purposes of malfunction detection have become unfeasible. As these systems output voluminous logs of their activity, machine led analysis of them is a growing need with already several existing solutions. These largely depend on having hand-crafted features, require raw log preprocessing and feature extraction or use supervised learning necessitating having a labeled log dataset not always easily procurable. We propose a two part deep autoencoder model with LSTM units that requires no hand-crafted features, no preprocessing of data as it works on raw text and outputs an anomaly score for each log entry. This anomaly score represents the rarity of a log event both in terms of its content and temporal context. The model was trained and tested on a dataset of HDFS logs containing 2 million raw lines of which half was used for training and half for testing. While this model cannot match the performance of a supervised binary classifier, it could be a useful tool as a coarse filter for manual inspection of log files where a labeled dataset is unavailable.

We present a generalization of the person-image generation task, in which a human image is generated conditioned on a target pose and a set X of source appearance images. In this way, we can exploit multiple, possibly complementary images of the same person which are usually available at training and at testing time. The solution we propose is mainly based on a local attention mechanism which selects relevant information from different source image regions, avoiding the necessity to build specific generators for each specific cardinality of X. The empirical evaluation of our method shows the practical interest of addressing the person-image generation problem in a multi-source setting.

This paper presents a novel strategy that employs Generative Adversarial Networks (GANs) to augment data in the skin lesion segmentation task, which is a fundamental first step in the automated melanoma detection process. The proposed framework generates both skin lesion images and their segmentation masks, making the data augmentation process extremely straightforward. In order to thoroughly analyze how the quality and diversity of synthetic images impact the efficiency of the method, we remodel two different well known GANs: a Deep Convolutional GAN (DCGAN) and a Laplacian GAN (LAPGAN). Experimental results reveal that, by introducing such kind of synthetic data into the training process, the overall accuracy of a state-of-the-art Convolutional/Deconvolutional Neural Network for melanoma skin lesion segmentation is increased.

Motivation: High-Throughput Next-Generation-Sequencing can generate huge sequence files, whose analysis requires alignment algorithms that are typically very demanding in terms of memory and computational resources. This is a significant issue, especially for machines with limited hardware capabilities. As the redundancy of the sequences typically increases with coverage, collapsing such files into compact sets of non-redundant reads has the two-fold advantage of reducing file size and speeding-up the alignment, avoiding to map the same sequence multiple times. Method: BioSeqZip generates compact and sorted lists of alignment-ready non-redundant sequences, keeping track of their occurrences in the raw files as well as of their quality score information. By exploiting a memory-constrained external sorting algorithm, it can be executed on either single or multi-sample data-sets even on computers with medium computational capabilities. On request, it can even re-expand the compacted files to their original state. Results: Our extensive experiments on RNA-seq data show that BioSeqZip considerably brings down the computational costs of a standard sequence analysis pipeline, with particular benefits for the alignment procedures that typically have the highest requirements in terms of memory and execution time. In our tests, BioSeqZip was able to compact 2.7 billions of reads into 963 millions of unique tags reducing the size of sequence files up to 70% and speeding-up the alignment by 50% at least. Availability: BioSeqZip is available at https://github.com/bioinformatics-polito/BioSeqZip Supplementary information: Supplementary data are available at Bioinformatics online.

Background/Aim: Proliferation biomarkers such as MIB-1 are strong predictors of clinical outcome and response to therapy in patients with non-small-cell lung cancer, but they require histological examination. In this work, we present a classification model to predict MIB-1 expression based on clinical parameters from positron emission tomography. Patients and Methods: We retrospectively evaluated 78 patients with histology-proven non-small-cell lung cancer (NSCLC) who underwent 18F-FDG-PET/CT for clinical examination. We stratified the population into a low and high proliferation group using MIB-1=25% as cut-off value. We built a predictive model based on binary classification trees to estimate the group label from the maximum standardized uptake value (SUVmax) and lesion diameter. Results: The proposed model showed ability to predict the correct proliferation group with overall accuracy >82% (78% and 86% for the low- and high-proliferation group, respectively). Conclusion: Our results indicate that radiotracer activity evaluated via SUVmax and lesion diameter are correlated with tumour proliferation index MIB-1.

Autonomous robotic platforms can be effectively used to perform automatic fruits yield estimation. To this aim, robots need data-driven models that process image streams and count, even approximately, the number of fruits in an orchard. However, training such models following a supervised paradigm is expensive and unpractical. Extending pre-trained models to perform yield estimation for a completely new type of fruit is even more challenging, but interesting since this situation is typical in practice. In this work, we combine a State-of-the-Art weakly-supervised fruit counting model with an unsupervised style transfer method for addressing the task above. In this sense, our proposed approach is quasi-unsupervised. In particular, we use a Cycle-Generative Adversarial Network (C-GAN) to perform unsupervised domain adaptation and train it alongside with a Presence-Absence Classifier (PAC) that discriminates images containing fruits or not. The PAC produces the weak-supervision signal for the counting network, that can then be used on the target orchard directly. Experiments on datasets collected in four different orchards show that the proposed approach is more accurate than the supervised baseline methods.

In this paper we present a novel approach for bottom-up multi-person 3D human pose estimation from monocular RGB images. We propose to use high resolution volumetric heatmaps to model joint locations, devising a simple and effective compression method to drastically reduce the size of this representation. At the core of the proposed method lies our Volumetric Heatmap Autoencoder, a fully-convolutional network tasked with the compression of ground-truth heatmaps into a dense intermediate representation. A second model, the Code Predictor, is then trained to predict these codes, which can be decompressed at test time to re-obtain the original representation. Our experimental evaluation shows that our method performs favorably when compared to state of the art on both multi-person and single-person 3D human pose estimation datasets and, thanks to our novel compression strategy, can process full-HD images at the constant runtime of 8 fps regardless of the number of subjects in the scene.