Publications

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.

Dorsal root ganglia (DRGs) host the somata of sensory neurons which convey information from the periphery to the central nervous system. These neurons have heterogeneous size and neurochemistry, and those of small-to-medium size, which play an important role in nociception, form two distinct subpopulations based on the presence (peptidergic) or absence (non-peptidergic) of transmitter neuropeptides. Few investigations have so far addressed the spatial relationship between neurochemically different subpopulations of DRG neurons and glia. We used a whole-mount mouse lumbar DRG preparation, confocal microscopy and computer-aided 3D analysis to unveil that IB4+ non-peptidergic neurons form small clusters of 4.7 ± 0.26 cells, differently from CGRP+ peptidergic neurons that are, for the most, isolated (1.89 ± 0.11 cells). Both subpopulations of neurons are ensheathed by a thin layer of satellite glial cells (SGCs) that can be observed after immunolabeling with the specific marker glutamine synthetase (GS). Notably, at the ultrastructural level we observed that this glial layer was discontinuous, as there were patches of direct contact between the membranes of two adjacent IB4+ neurons. To test whether this cytoarchitectonic organization was modified in the diabetic neuropathy, one of the most devastating sensory pathologies, mice were made diabetic by streptozotocin (STZ). In diabetic animals, cluster organization of the IB4+ non-peptidergic neurons was maintained, but the neuro-glial relationship was altered, as STZ treatment caused a statistically significant increase of GS staining around CGRP+ neurons but a reduction around IB4+ neurons. Ultrastructural analysis unveiled that SGC coverage was increased at the interface between IB4+ cluster-forming neurons in diabetic mice, with a 50% reduction in the points of direct contacts between cells. These observations demonstrate the existence of a structural plasticity of the DRG cytoarchitecture in response to STZ.

Continual Learning has inspired a plethora of approaches and evaluation settings; however, the majority of them overlooks the properties of a practical scenario, where the data stream cannot be shaped as a sequence of tasks and offline training is not viable. We work towards General Continual Learning (GCL), where task boundaries blur and the domain and class distributions shift either gradually or suddenly. We address it through mixing rehearsal with knowledge distillation and regularization; our simple baseline, Dark Experience Replay, matches the network's logits sampled throughout the optimization trajectory, thus promoting consistency with its past. By conducting an extensive analysis on both standard benchmarks and a novel GCL evaluation setting (MNIST-360), we show that such a seemingly simple baseline outperforms consolidated approaches and leverages limited resources. We further explore the generalization capabilities of our objective, showing its regularization being beneficial beyond mere performance.

Nowadays, robots are heavily used in factories for different tasks, most of them including grasping and manipulation of generic objects in unstructured scenarios. In order to better mimic a human operator involved in a grasping action, where he/she needs to identify the object and detect an optimal grasp by means of visual information, a widely adopted sensing solution is Artificial Vision. Nonetheless, state-of-art applications need long training and fine-tuning for manually build the object's model that is used at run-time during the normal operations, which reduce the overall operational throughput of the robotic system. To overcome such limits, the paper presents a framework based on Deep Convolutional Neural Networks (DCNN) to predict both single and multiple grasp poses for multiple objects all at once, using a single RGB image as input. Thanks to a novel loss function, our framework is trained in an end-to-end fashion and matches state-of-art accuracy with a substantially smaller architecture, which gives unprecedented real-time performances during experimental tests, and makes the application reliable for working on real robots. The system has been implemented using the ROS framework and tested on a Baxter collaborative robot.

Summary: In the last decade, increasing attention has been paid to the study of gene fusions. However, the problem of determining whether a gene fusion is a cancer driver or just a passenger mutation is still an open issue. Here we present DEEPrior, an inherently flexible deep learning tool with two modes (Inference and Retraining). Inference mode predicts the probability of a gene fusion being involved in an oncogenic process, by directly exploiting the amino acid sequence of the fused protein. Retraining mode allows to obtain a custom prediction model including new data provided by the user. Availability and implementation: Both DEEPrior and the protein fusions dataset are freely available from GitHub at (https://github.com/bioinformatics-polito/DEEPrior). The tool was designed to operate in Python 3.7, with minimal additional libraries. Supplementary information: Supplementary data are available at Bioinformatics online.

We address the problem of unsupervised gaze correction in the wild, presenting a solution that works without the need of precise annotations of the gaze angle and the head pose. We created a new dataset called CelebAGaze consisting of two domains X, Y, where the eyes are either staring at the camera or somewhere else. Our method consists of three novel modules: the Gaze Correction module(GCM), the Gaze Animation module(GAM), and the Pretrained Autoencoder module (PAM). Specifically, GCM and GAM separately train a dual in-painting network using data from the domain X for gaze correction and data from the domain Y for gaze animation. Additionally, a Synthesis-As-Training method is proposed when training GAM to encourage the features encoded from the eye region to be correlated with the angle information, resulting in gaze animation achieved by interpolation in the latent space. To further preserve the identity information e.g., eye shape, iris color, we propose the PAM with an Autoencoder, which is based on Self-Supervised mirror learning where the bottleneck features are angle-invariant and which works as an extra input to the dual in-painting models. Extensive experiments validate the effectiveness of the proposed method for gaze correction and gaze animation in the wild and demonstrate the superiority of our approach in producing more compelling results than state-of-the-art baselines. Our code, the pretrained models and supplementary results are available at:https://github.com/zhangqianhui/GazeAnimation.

Clustering methods are increasingly applied to single-cell DNA sequencing (scDNAseq) data to infer the subclonal structure of cancer. However, the complexity of these data exacerbates some data-science issues and affects clustering results. Additionally, determining whether such inferences are accurate and clusters recapitulate the real cell phylogeny is not trivial, mainly because ground truth information is not available for most experimental settings. Here, by exploiting simulated sequencing data representing known phylogenies of cancer cells, we propose a formal and systematic assessment of well-known clustering methods to study their performance and identify the approach providing the most accurate reconstruction of phylogenetic relationships.