Search

Deep learning approaches have recently raised the bar in many fields, from Natural Language Processing to Computer Vision, by leveraging large amounts of data. However, they could fail when the retrieved information is not enough to fit the vast number of parameters, frequently resulting in overfitting and therefore in poor generalizability. Few-Shot Learning aims at designing models that can effectively operate in a scarce data regime, yielding learning strategies that only need few supervised examples to be trained. These procedures are of both practical and theoretical importance, as they are crucial for many real-life scenarios in which data is either costly or even impossible to retrieve. Moreover, they bridge the distance between current data-hungry models and human-like generalization capability. Computer vision offers various tasks that can be few-shot inherent, such as person re-identification. This survey, which to the best of our knowledge is the first tackling this problem, is focused on Few-Shot Object Detection, which has received far less attention compared to Few-Shot Classification due to the intrinsic challenge level. In this regard, this review presents an extensive description of the approaches that have been tested in the current literature, discussing their pros and cons, and classifying them according to a rigorous taxonomy. [ABSTRACT FROM AUTHOR]

In this article, we explore the correlation between people trajectories and their head orientations. We argue that people trajectory and head pose forecasting can be modelled as a joint problem. Recent approaches on trajectory forecasting leverage short-term trajectories (aka tracklets) of pedestrians to predict their future paths. In addition, sociological cues, such as expected destination or pedestrian interaction, are often combined with tracklets. In this article, we propose MiXing-LSTM (MX-LSTM) to capture the interplay between positions and head orientations (vislets) thanks to a joint unconstrained optimization of full covariance matrices during the LSTM backpropagation. We additionally exploit the head orientations as a proxy for the visual attention, when modeling social interactions. MX-LSTM predicts future pedestrians location and head pose, increasing the standard capabilities of the current approaches on long-term trajectory forecasting. Compared to the state-of-the-art, our approach shows better performances on an extensive set of public benchmarks. MX-LSTM is particularly effective when people move slowly, i.e., the most challenging scenario for all other models. The proposed approach also allows for accurate predictions on a longer time horizon. [ABSTRACT FROM AUTHOR]

Simple Summary: Prostate-specific antigen and digital rectal examination, used to guide prostate biopsy, often result in overdiagnosis of indolent prostate cancer (PCa) while missing clinically significant PCa (csPCa). The aim of this study was to evaluate the diagnostic accuracy of SelectMDx and its association with multiparametric magnetic resonance imaging (mpMRI) in predicting PCa in prostate biopsies. SelectMDx was revealed to be a good predictor of PCa, while with regards to csPCa detection, it was demonstrated to be less effective, showing results similar to mpMRI. The best diagnostic strategy to avoid unnecessary biopsy is performing SelectMDx after an initial negative mpMRI. Biopsy could be proposed for all cases of positive mpMRI and to those with a negative mpMRI followed by a positive SelectMDx. Prostate-specific antigen (PSA) testing as the sole indication for prostate biopsy lacks specificity, resulting in overdiagnosis of indolent prostate cancer (PCa) and missing clinically significant PCa (csPCa). SelectMDx is a biomarker-based risk score to assess urinary HOXC6 and DLX1 mRNA expression combined with traditional clinical risk factors. The aim of this prospective multi-institutional study was to evaluate the diagnostic accuracy of SelectMDx and its association with multiparametric magnetic resonance (mpMRI) when predicting PCa in prostate biopsies. Overall, 310 consecutive subjects were included. All patients underwent mpMRI and SelectMDx prior to prostate biopsy. SelectMDx and mpMRI showed sensitivity and specificity of 86.5% vs. 51.9%, and 73.8% vs. 88.3%, respectively, in predicting PCa at biopsy, and 87.1% vs. 61.3%, and 63.7% vs. 83.9%, respectively, in predicting csPCa at biopsy. SelectMDx was revealed to be a good predictor of PCa, while with regards to csPCa detection, it was demonstrated to be less effective, showing results similar to mpMRI. With analysis of strategies assessed to define the best diagnostic strategy to avoid unnecessary biopsy, SelectMDx appeared to be a reliable pathway after an initial negative mpMRI. Thus, biopsy could be proposed for all cases of mpMRI PI-RADS 4–5 score, and to those with Prostate Imaging-Reporting and Data System (PI-RADS) 1–3 score followed by a positive SelectMDx. [ABSTRACT FROM AUTHOR]

Deep-learning models for 3D point cloud semantic segmentation exhibit limited generalization capabilities when trained and tested on data captured with different sensors or in varying environments due to domain shift. Domain adaptation methods can be employed to mitigate this domain shift, for instance, by simulating sensor noise, developing domain-agnostic generators, or training point cloud completion networks. Often, these methods are tailored for range view maps or necessitate multi-modal input. In contrast, domain adaptation in the image domain can be executed through sample mixing, which emphasizes input data manipulation rather than employing distinct adaptation modules. In this study, we introduce compositional semantic mixing for point cloud domain adaptation, representing the first unsupervised domain adaptation technique for point cloud segmentation based on semantic and geometric sample mixing. We present a two-branch symmetric network architecture capable of concurrently processing point clouds from a source domain (e.g. synthetic) and point clouds from a target domain (e.g. real-world). Each branch operates within one domain by integrating selected data fragments from the other domain and utilizing semantic information derived from source labels and target (pseudo) labels. Additionally, our method can leverage a limited number of human point-level annotations (semi-supervised) to further enhance performance. We assess our approach in both synthetic-to-real and real-to-real scenarios using LiDAR datasets and demonstrate that it significantly outperforms state-of-the-art methods in both unsupervised and semi-supervised settings.

Introduction: Chronic benign prostate diseases are very common and certainly feature significantly in urological practice.The treatment of chronic benign prostate diseases is a common problem in clinical practice: few studies have been conducted in routine clinical practice to evaluate the efficacy of the treatments for this clinical condition. The objective of this study was to evaluate the efficacy of an extract of Serenoa repens (Permixon) in the treatment of lower urinary tract symptoms (LUTS) in patients with chronic benign prostate diseases with associated inflammation, also taking into consideration the influence of treatment on sexual function and, therefore, on patients' quality of life., Materials and Methods: All the 591 eligible subjects were evaluated on entering the study; after a screening visit, including medical history, physical examination, physical examination and digital rectal examination (DRE) and laboratory tests, the patients underwent uroflowmetry. The subjects under investigation were also asked to complete the IPSS, NIH-CPSI and IIEF-5 questionnaires, for the purpose of evaluating urinary symptoms and erectile function in relation to sexual activity in the previous 6 months., Results: The analysis of the uroflowmetry results showed that treatment with extract of Serenoa repens distinctly improves bladder voiding and lower urinary tract symptoms, as highlighted also by the improvement in the scores for the IPSS and NIH-CPSI questionnaires which serve as a basis for evaluating the urinary symptoms of patients with prostatic hyperplasia and chronic prostatitis respectively. The results also suggest that using an extract of Serenoa repens for 6 months in patients with chronic benign prostate diseases gives rise to an improvement in erectile function, as demonstrated by the increase in the scores for the IIEF-5 questionnaire after 6 months of treatment., Conclusions: The results of this study demonstrate how treatment for 6 months with an extract of Serenoa repens in routine clinical practice gives rise to a statistically significant improvement in Qmax values and in the IPSS, NHI-CPSI and IIEF-5 questionnaire scores, resulting not only in an improvement in urinary symptoms but also in an overall improvement in patients' quality of life.

Objectives: PCA3 is a prostate specific non-coding mRNA that is significantly overexpressed in prostate cancer tissue. Urinary PCA3 levels have been associated with prostate cancer grade suggesting a significant role in the diagnosis of prostate cancer. We measured urinary PCA3 score in 925 subjects from several areas of Italy assessing in 114 the association of urinary PCA3 score with the results of prostate biopsy., Material and Methods: First-catch urine samples were collected after digital rectal examination (DRE). PCA3 and PSA mRNA levels were measured using Trascription-mediated PCR amplification. The PCA3 score was calculated as the ratio of PCA3 and PSA mRNA (PCA3 mRNA/PSA mRNA x 1000) and the cut off was set at 35., Results: A total of 925 PCA3 tests were performed from December 2008 to January 2010. The rate of informative PCA3 test was 99%, with 915 subjects showing a valid PCA3 score value: 443 patients (48.42%) presented a PCA3 score >/= 35 (cut-off) whereas the remaining 472 patients (51.58%) presented a PCA3 score lower the cut-off limit (< 35). Of the 443 patients with PCA3 score >/= 35, 105 (23.70%) underwent biopsy or rebiopsy. We found that 27 patients (25.71%) had no tumour at biopsy, 37 (35.24%) had HGPIN or ASAP and 41 (39.05%) had a cancer. Moreover, including the additonal 9 patients with PCA3 < 35, who underwent biopsy post PCA3 results, our data indicate that patients with negative biopsy (n = 31) show lower PCA3 score (mean = 54.9) compared with patients with positive biopsy (n = 45) (mean = 141.6) (p = 0.000183; two-tailed t-student test). The mean PCA3 score (79.6)for the patients diagnosed with HGPIN/ASAP at biopsy (n = 38) was intermediate between patients with negative and positive biopsy., Conclusions: Our results indicate that the PCA3 score is a valid tool for prostate cancer detection and its role in making better biopsy decisions. This marker consents to discriminate patients who have to undergo biopsy from patients who only need be actively surveilled: Quantitative PCA3 score is correlated with the probability of a positive result at biopsy.