Search

Simple Summary: Thyroid nodules must be accurately classified as benign or malignant. The aim of this study is to develop a machine learning model based on thyroid ultrasound images in order to classify nodules into the two classes. Ultrasound (US) images were collected from 142 patients for training, validation and internal testing of three models, plus 21 images to externally test the best performing model. The random forest classifier model could perform the classification task, identifying all the malignant nodes and most of the benign. Background/Objectives: Thyroid nodules are a very common finding, mostly benign but sometimes malignant, and thus require accurate diagnosis. Ultrasound and fine needle biopsy are the most widely used and reliable diagnostic methods to date, but they are sometimes limited in addressing benign from malignant nodules, mainly with regard to ultrasound, by the operator's experience. Radiomics, quantitative feature extraction from medical images and machine learning offer promising avenues to improve diagnosis. The aim of this work was to develop a machine learning model based on thyroid ultrasound images to classify nodules into benign and malignant classes. Methods: For this purpose, images of ultrasonography from 142 subjects were collected. Among these subjects, 40 patients (28.2%) belonged to the class "malignant" and 102 patients (71.8%) belonged to the class "benign", according to histological diagnosis from fine-needle aspiration. This image set was used for the training, cross-validation and internal testing of three different machine learning models. A robust radiomic approach was applied, under the hypothesis that the radiomic feature could capture the disease heterogeneity among the two groups. Three models consisting of four ensembles of machine learning classifiers (random forests, support vector machines and k-nearest neighbor classifiers) were developed for the binary classification task of interest. The best performing model was then externally tested on a cohort of 21 new patients. Results: The best model (ensemble of random forest) showed Receiver Operating Characteristic-Area Under the Curve (ROC-AUC) (%) of 85 (majority vote), 83.7 ** (mean) [80.2–87.2], accuracy (%) of 83, 81.2 ** [77.1–85.2], sensitivity (%) of 70, 67.5 ** [64.3–70.7], specificity (%) of 88, 86.5 ** [82–91], positive predictive value (PPV) (%) of 70, 66.5 ** [57.9–75.1] and negative predictive value (NPV) (%) of 88, 87.1 ** [85.5–88.8] (* p < 0.05, ** p < 0.005) in the internal test cohort. It achieved an accuracy of 90.5%, a sensitivity of 100%, a specificity of 86.7%, a PPV of 75% and an NPV of 100% in the external testing cohort. Conclusions: The model constituted of four ensembles of random forest classifiers could identify all the malignant nodes and the consistent majority of benign in the external testing cohort. [ABSTRACT FROM AUTHOR]

Background: About 20% of hereditary breast cancers are caused by mutations in BRCA1 and BRCA2 genes. Since BRCA1 and BRCA2 mutations may be spread throughout the gene, genetic testing is usually performed by direct sequencing of entire coding regions. In some populations, especially if relatively isolated, a few number of recurrent mutations is reported, sometimes caused by founder effect. Methods: BRCA1 and BRCA2 screening for mutations was carried out on 1114 breast and/or ovarian cancer patients complying with the eligibility criteria for BRCA testing. Haplotype analysis was performed on the probands carrying recurrent mutations and their relatives, using two sets of microsatellite markers covering the BRCA1 (D17S588, D17S806, D17S902, D17S1325, D17S855, D17S1328, D17S800, and D17S250) and BRCA2 (D13S220, D13S267, D13S171, D13S1701, D13S1698, D13S260, D13S290, D13S1246) loci. The DMLE + 2.2 software was used to estimate the age of BRCA1 c.676delT and BRCA2 c.7806-2A > G. A multiplex PCR and two different primer extension assays were optimized and used for genotyping the recurrent mutations of the two genes. Results: In the time frame of almost 20 years of genetic testing, we have found that five BRCA1 and three BRCA2 mutations are recurrent in a substantial subset of carriers from North-East Italy and neighboring Istria, where they represent more than 50% of all mutations. Microsatellite analyses identified a common haplotype of different length for each mutation. Age estimation of BRCA1 c.676delT and BRCA2 c.7806-2A > G mutations revealed that they arose in the Friuli Venezia Giulia area about 86 and 94 generations ago, respectively. Suggestion of an association between BRCA2 c.7806-2A > G and risk of breast cancer in males has emerged. Finally, we developed a simple and efficient pre-screeening test, performing an in-house primer extension SNaPshot® assay for the rapid identification of the eight recurrent mutations. Conclusions: Proofs of common ancestry has been obtained for the eight recurrent mutations. The observed genotype-phenotype correlation and the proposed rapid mutation detection strategy could improve the clinical management of breast and ovarian patients in North-East of Italy and neighboring geographic areas. [ABSTRACT FROM AUTHOR]

Background: Breast cancer in men is an infrequent occurrence, accounting for ~1% of all breast tumors with an incidence of about 1:100,000. The relative rarity of male breast cancer (MBC) limits our understanding of the epidemiologic, genetic and clinical features of this tumor. Methods: From 1997 to 2003, 10 MBC patients were referred to our Institute for genetic counselling and BRCA1/2 testing. Here we report on the genetic and phenotypic characterization of 10 families with MBC from the North East of Italy. In particular, we wished to assess the occurrence of specific cancer types in relatives of MBC probands in families with and without BRCA2 predisposing mutations. Moreover, families with recurrent BRCA2 mutations were also characterized by haplotype analysis using 5 BRCA2-linked dinucleotide repeat markers and 8 intragenic BRCA2 polymorphisms. Results: Two pathogenic mutations in the BRCA2 gene were observed: the 9106C>T (Q2960X) and the IVS16-2A>G (splicing) mutations, each in 2 cases. A BRCA1 mutation of uncertain significance 4590C>G (P1491A) was also observed. In families with BRCA2 mutations, female breast cancer was more frequent in the first and second-degree relatives compared to the families with wild type BRCA1/2 (31.9% vs. 8.0% p = 0.001). Reconstruction of the chromosome phasing in three families and the analysis of three isolated cases with the IVS16-2A>G BRCA2 mutation identified the same haplotype associated with MBC, supporting the possibility that this founder mutation previously detected in Slovenian families is also present in the North East of our Country. Moreover, analysis of one family with the 9106C>T BRCA2 mutation allowed the identification of common haplotypes for both microsatellite and intragenic polymorphisms segregating with the mutation. Three isolated cases with the same mutation shared the same intragenic polymorphisms and three 5' microsatellite markers, but showed a different haplotype for 3' markers, which were common to all three cases. Conclusion: The 9106C>T and the IVS16-2A>G mutations constitute recurrent BRCA2 mutations in MBC cases from the North-East of Italy and may be associated with a founder effect. Knowledge of these two recurrent BRCA2 mutations predisposing to MBC may facilitate the analyses aimed at the identification of mutation carriers in our geographic area [ABSTRACT FROM AUTHOR]

Drinkable water supplied by aqueducts undergoes preliminar potabilization which, in Italy, is mainly accomplished by chlorine addition. The bactericidal action involved in this process is always accompanied by chlorination and oxidation of organic species (mainly humic and fulvic acids) naturally present in treated waters, so that many disinfection by-products (DBPs) are formed, such as trihalomethanes (THMs) and halo-acetic acids (HAA), which can represent a chemical risk for public health. The aim of this study was the monitoring of DBPs in drinking water disinfected by chlorination, supplied by four different aqueducts of Central Friuli (Italy). DBP evaluations were performed in water samples consisting of both input and output of disinfection plants. The results of analytical determinations were worked out to provide the THM and HAA parameters for disinfected waters, while in feeding waters the following different conventional parameters were adopted : (1) trihalomethanes formation potential (THMFP), (ii) halo-acetic acids formation potential (HAAFP) and (iii) UV absorbance at 254 nm (UV254). The quite moderate content of chlorinated products found in all samples considered highlighted the excellent quality of potabilized waters available in Central Friuli. Moreover, our results confirmed that the majority of DBPs formed when chlorine is used for water disinfection consists of THMs, while chlorites and chlorates prevailed when potabilization is accomplished by using chlorine dioxide. Finally, simple UV254 monitoring turned out to be a profitable approach for the determination of chlorinated by-products only when THMs prevail among DBPs. [ABSTRACT FROM AUTHOR]

Erdosteine is a new thiol compound with effects on bacterial adhesiveness as well as antioxidant and mucoactive properties. The EQUALIFE study, a fully randomized, double-blind, placebo-controlled, parallel-group, multicenter study, was designed to assets the effectiveness of long-term treatment with erdosteine in patients with moderate chronic obstructive pulmonary disease (COPD). One hundred and fifty-five patients received oral erdosteine, 300 mg b.i.d., or placebo for 8 months during the winter season to assess the effect of treatments on exacerbation rate, hospitalization, lung function and quality of life, assessed using the Short Form 36 and the St. George's Respiratory Questionnaire. A pharmacoeconomic analysis was also conducted to compare the two treatments. One hundred and twenty-four patients completed the study with erdosteine (n = 63) or placebo (n = 61). The group of COPD patients who received 8 months of continuous treatment with erdosteine had significantly fewer exacerbations and spent fewer days in the hospital than did the placebo group; furthermore, they had no loss of lung function. Patients in the erdosteine group also showed a significant improvement in health-related quality of life. The mean total COPD-related disease costs per patient were lower in the erdosteine group than in the placebo group over the study period. The results indicate that 8 months of treatment with erdosteine is effective in reducing exacerbation and hospitalization rates and in improving health status. The study suggests that erdosteine is likely to provide an important contribution to the therapy of patients with symptomatic COPD.