Background: The occurrence of spontaneous tumors in pet animals has been estimated in a few European and North American veterinary cancer registries with dissimilar methodologies and variable reference populations.Objectives: The Animal Tumor Registry (ATR) of Genoa, Italy, was established in 1985 with the aim of estimating the occurrence of spontaneous tumors in dogs.Methods: Six thousand seven hundred and forty-three tumor biopsy specimens were received from local veterinarians in the Municipality of Genoa between 1985 and 2002. Three thousand and three hundred and three (48.9%) biopsy specimen samples were diagnosed as cancer and were coded according to the International Statistical Classification of Diseases (ICD-9).Results: Mammary cancer was the most frequently diagnosed cancer in female dogs, accounting for 70% of all cancer cases. Incidence of all cancers was 99.3 per 100,000 dog-years (95% CI: 93.6-105.1) in male dogs and 272.1 (95% CI: 260.7-283.6) in female dogs. The highest incidence rates were detected for mammary cancer (IR 5 191.8, 95% CI: 182.2-201.4) and for non-Hodgkin's lymphoma (IR 5 22.9, 95% CI: 19.7-26.5) in bitches and for non-Hodgkin's lymphoma (IR 5 19.9, 95% CI: 17.4-22.7) and skin cancer (IR 5 19.1, 95% CI: 16.6-21.8) in male dogs. All cancer IR increased with age ranging between 23.7 (95% CI: 18.4-30.1) and 763.2 (95% CI: 700.4-830.1) in bitches and between 16.5 (95% CI: 12.8-21.1) and 237.6 (95% CI: 209.1-269.0) in male dogs aged 3 years and 49-11 years.Conclusion: This study summarizes the work done by the ATR of Genoa, Italy, between 1985 and 2002. All cancer incidence was 3 times higher in female than in male dogs, a difference explained by the high rate of mammary cancer observed in bitches. Because a biopsy specimen was required to make a cancer diagnosis, cancer rates for internal organs cancers, such as respiratory and digestive tract cancers may have been underestimated in the study population.

No abstract

No abstract

Abstract-In this work we address the problem of estimating parameters of diffusion phenomena via autonomous wireless sensor networks. Diffusion phenomena, such as the propagation of a gas in the air or of a chemical agent in the water, can be modeled by means of partial differential equations (PDE's). In several scenarios, the parameters characterizing such models, i.e. the coefficients of the PDE's, are not known a-priori and need to be estimated. We develop an adaptive approach for the distributed identification of the parameters of diffusion models for both the cases of known and unknown boundary conditions (BC's). The technique also applies to the case of spatially varying parameters. We present simulation results to show the performance and the various trade-offs of the method.

No abstract

No abstract

Background and Objectives Post‐discharge oncologic surgical complications are costly for patients, families, and healthcare systems. The capacity to predict complications and early intervention can improve postoperative outcomes. In this proof‐of‐concept study, we used a machine learning approach to explore the potential added value of patient‐reported outcomes (PROs) and patient‐generated health data (PGHD) in predicting post‐discharge complications for gastrointestinal (GI) and lung cancer surgery patients. Methods We formulated post‐discharge complication prediction as a binary classification task. Features were extracted from clinical variables, PROs (MD Anderson Symptom Inventory [MDASI]), and PGHD (VivoFit) from a cohort of 52 patients with 134 temporal observation points pre‐ and post‐discharge that were collected from two pilot studies. We trained and evaluated supervised learning classifiers via nested cross‐validation. Results A logistic regression model with L2 regularization trained with clinical data, PROs and PGHD from wearable pedometers achieved an area under the receiver operating characteristic of 0.74. Conclusions PROs and PGHDs captured through remote patient telemonitoring approaches have the potential to improve prediction performance for postoperative complications.

Key Points Question How do oncologists and a machine learning model compare in predicting 3-month mortality for patients with advanced solid tumors? Findings In this prognostic study, the machine learning model significantly outperformed 74 oncologists in predicting 3-month mortality for 2041 patients with metastatic solid tumors overall and in gastrointestinal and breast cancer subpopulations. Findings were not significant in genitourinary, lung, and rare cancer groups. Meaning The results of this study suggest the potential for a machine learning model trained with electronic health record data to support oncologists in prognostication and clinical decision-making to improve end-of-life care.