Crosslingual named entity recognition for clinical de-identification applied to a COVID-19 Italian data set.

The COrona VIrus Disease 19 (COVID-19) pandemic required the work of all global experts to tackle it. Despite the abundance of new studies, privacy laws prevent their dissemination for medical investigations: through clinical de-identification, the Protected Health Information (PHI) contained therein can be anonymized so that medical records can be shared and published. The automation of clinical de-identification through deep learning techniques has proven to be less effective for languages other than English due to the scarcity of data sets. Hence a new Italian de-identification data set has been created from the COVID-19 clinical records made available by the Italian Society of Radiology (SIRM). Therefore, two multi-lingual deep learning systems have been developed for this low-resource language scenario: the objective is to investigate their ability to transfer knowledge between different languages while maintaining the necessary features to correctly perform the Named Entity Recognition task for de-identification. The systems were trained using four different strategies, using both the English Informatics for Integrating Biology & the Bedside (i2b2) 2014 and the new Italian SIRM COVID-19 data sets, then evaluated on the latter. These approaches have demonstrated the effectiveness of cross-lingual transfer learning to de-identify medical records written in a low resource language such as Italian, using one with high resources such as English.

Best practices for portfolio optimization by quantum computing, experimented on real quantum devices.

In finance, portfolio optimization aims at finding optimal investments maximizing a trade-off between return and risks, given some constraints. Classical formulations of this quadratic optimization problem have exact or heuristic solutions, but the complexity scales up as the market dimension increases. Recently, researchers are evaluating the possibility of facing the complexity scaling issue by employing quantum computing. In this paper, the problem is solved using the Variational Quantum Eigensolver (VQE), which in principle is very efficient. The main outcome of this work consists of the definition of the best hyperparameters to set, in order to perform Portfolio Optimization by VQE on real quantum computers. In particular, a quite general formulation of the constrained quadratic problem is considered, which is translated into Quadratic Unconstrained Binary Optimization by the binary encoding of variables and by including constraints in the objective function. This is converted into a set of quantum operators (Ising Hamiltonian), whose minimum eigenvalue is found by VQE and corresponds to the optimal solution. In this work, different hyperparameters of the procedure are analyzed, including different ansatzes and optimization methods by means of experiments on both simulators and real quantum computers. Experiments show that there is a strong dependence of solutions quality on the sufficiently sized quantum computer and correct hyperparameters, and with the best choices, the quantum algorithm run on real quantum devices reaches solutions very close to the exact one, with a strong convergence rate towards the classical solution, even without error-mitigation techniques. Moreover, results obtained on different real quantum devices, for a small-sized example, show the relation between the quality of the solution and the dimension of the quantum processor. Evidences allow concluding which are the best ways to solve real Portfolio Optimization problems by VQE on quantum devices, and confirm the possibility to solve them with higher efficiency, with respect to existing methods, as soon as the size of quantum hardware will be sufficiently high.

A Novel COVID-19 Data Set and an Effective Deep Learning Approach for the De-Identification of Italian Medical Records.

In the last years, the need to de-identify privacy-sensitive information within Electronic Health Records (EHRs) has become increasingly felt and extremely relevant to encourage the sharing and publication of their content in accordance with the restrictions imposed by both national and supranational privacy authorities. In the field of Natural Language Processing (NLP), several deep learning techniques for Named Entity Recognition (NER) have been applied to face this issue, significantly improving the effectiveness in identifying sensitive information in EHRs written in English. However, the lack of data sets in other languages has strongly limited their applicability and performance evaluation. To this aim, a new de-identification data set in Italian has been developed in this work, starting from the 115 COVID-19 EHRs provided by the Italian Society of Radiology (SIRM): 65 were used for training and development, the remaining 50 were used for testing. The data set was labelled following the guidelines of the i2b2 2014 de-identification track. As additional contribution, combined with the best performing Bi-LSTM + CRF sequence labeling architecture, a stacked word representation form, not yet experimented for the Italian clinical de-identification scenario, has been tested, based both on a contextualized linguistic model to manage word polysemy and its morpho-syntactic variations and on sub-word embeddings to better capture latent syntactic and semantic similarities. Finally, other cutting-edge approaches were compared with the proposed model, which achieved the best performance highlighting the goodness of the promoted approach.

Indoor Radon Concentration and Risk Assessment in 27 Districts of a Public Healthcare Company in Naples, South Italy.

Radon is a major source of ionizing radiation exposure for the general population. It is known that exposure to radon is a risk factor for the onset of lung cancer. In this study, the results of a radon survey conducted in all districts of a Public Healthcare in Italy, are reported. Measurements of indoor radon were performed using nuclear track detectors, CR-39. The entire survey was conducted according to a well-established quality assurance program. The annual effective dose and excess lifetime cancer risk were also calculated. Results show that the radon concentrations varied from 7 ± 1 Bq/m3 and 5148 ± 772 Bq/m3, with a geometric mean of 67 Bq/m3 and geometric standard deviation of 2.5. The annual effective dose to workers was found to be 1.6 mSv/y and comparable with the worldwide average. In Italy, following the transposition of the European Directive 59/2013, great attention was paid to the radon risk in workplaces. The interest of the workers of the monitored sites was very high and this, certainly contributed to the high return rate of the detectors after exposure and therefore, to the presence of few missing data. Although it was not possible to study the factors affecting radon concentrations, certainly the main advantage of this study is that it was the first in which an entire public health company was monitored in regards to all the premises on the underground and ground floor.

Subject identification via ECG fiducial-based systems: influence of the type of QT interval correction.

Electrocardiography (ECG) has been recently proposed as biometric trait for identification purposes. Intra-individual variations of ECG might affect identification performance. These variations are mainly due to Heart Rate Variability (HRV). In particular, HRV causes changes in the QT intervals along the ECG waveforms. This work is aimed at analysing the influence of seven QT interval correction methods (based on population models) on the performance of ECG-fiducial-based identification systems. In addition, we have also considered the influence of training set size, classifier, classifier ensemble as well as the number of consecutive heartbeats in a majority voting scheme. The ECG signals used in this study were collected from thirty-nine subjects within the Physionet open access database. Public domain software was used for fiducial points detection. Results suggested that QT correction is indeed required to improve the performance. However, there is no clear choice among the seven explored approaches for QT correction (identification rate between 0.97 and 0.99). MultiLayer Perceptron and Support Vector Machine seemed to have better generalization capabilities, in terms of classification performance, with respect to Decision Tree-based classifiers. No such strong influence of the training-set size and the number of consecutive heartbeats has been observed on the majority voting scheme.

A study on textual features for medical records classification.

Healthcare domain is characterized by a huge amount of data, contained in medical records, reports, test results and so on. In order to give support to healthcare workers and manage relevant data in effective and efficient way, it is important to correctly classify the unstructured parts of text, embedded in the medical documents. In this paper, we propose a classification system for medical records categorization, focused on the combination of different methodologies, based on lexical, syntactical and semantic analysis of the documents. We will show that a Classification System based on a combination of different text analysis methodologies overcomes the performances of each methodology taken alone. The obtained results will be presented in terms of Accuracy-Rejection Curves. Eventually, pro and cons of the architecture proposed and some future work will be pointed out.

Lifestyle educational program strongly increases compliance to nonpharmacologic intervention in hypertensive patients: a 2-year follow-up study.

The authors investigated the efficacy of a lifestyle educational program, organized in small group meetings, in improving the outcome of a nonpharmacologic intervention. One hundred and eighty-eight hypertensive patients with stable blood pressure (BP) levels and drug therapy in the previous 6 months were randomly divided into educational care (EC) and usual care (UC) groups. They were followed at 3-month intervals up to 2 years. In addition to the visits in an outpatient clinic, patients in the EC program participated in small group meetings in order to improve their knowledge of the disease and reinforce their motivation for treatment. At baseline, EC and UC groups were similar for age, sex, body mass index (BMI), blood pressure (BP) levels, and pharmacologic treatment. Patients in the EC group had significantly reduced total energy, total and saturated fats, and sodium intake. Physical activity was significantly increased in the EC group as well. At the end of the 1-year follow-up, BMI (P<.001), visceral fat (P<.001), and BP (P<.001) were significantly lower in the EC group compared with the UC group. Pharmacologic treatment during the study was similar for all classes of drugs apart from diuretics whose dose was higher in the UC group at the end of the study.