Radiomics from Various Tumour Volume Sizes for Prognosis Prediction of Head and Neck Squamous Cell Carcinoma: A Voted Ensemble Machine Learning Approach.

Background: Traditionally, cancer prognosis was determined by tumours size, lymph node spread and presence of metastasis (TNM staging). Radiomics of tumour volume has recently been used for prognosis prediction. In the present study, we evaluated the effect of various sizes of tumour volume. A voted ensemble approach with a combination of multiple machine learning algorithms is proposed for prognosis prediction for head and neck squamous cell carcinoma (HNSCC). Methods: A total of 215 HNSCC CT image sets with radiotherapy structure sets were acquired from The Cancer Imaging Archive (TCIA). Six tumour volumes, including gross tumour volume (GTV), diminished GTV, extended GTV, planning target volume (PTV), diminished PTV and extended PTV were delineated. The extracted radiomics features were analysed by decision tree, random forest, extreme boost, support vector machine and generalized linear algorithms. A voted ensemble machine learning (VEML) model that optimizes the above algorithms was used. The receiver operating characteristic area under the curve (ROC-AUC) were used to compare the performance of machine learning methods, including accuracy, sensitivity and specificity. Results: The VEML model demonstrated good prognosis prediction ability for all sizes of tumour volumes with reference to GTV and PTV with high accuracy of up to 88.3%, sensitivity of up to 79.9% and specificity of up to 96.6%. There was no significant difference between the various target volumes for the prognostic prediction of HNSCC patients (chi-square test, p > 0.05). Conclusions: Our study demonstrates that the proposed VEML model can accurately predict the prognosis of HNSCC patients using radiomics features from various tumour volumes.

Occurrence of Potential Adverse Drug Events from Prescribing Errors in a Pediatric Intensive and High Dependency Unit in Hong Kong: An Observational Study.

BACKGROUND: Critically ill pediatric patients are considered at high risk for medication errors. Although much research focuses on the actual errors, equally important are medication errors that, although intercepted, carried the potential for an adverse drug event. The aim of this study was to determine the occurrence of prescribing errors and potential adverse drug events (pADEs) in a local pediatric intensive and critical care unit (PICU) in Hong Kong. Our secondary objective was to determine the type of error, nature of medication involved and the time of error occurrence. METHODS: We conducted a prospective observational chart review among patients in a pediatric intensive and high dependency unit between January 16, 2015 and April 20, 2015. Medical charts for each patient were reviewed for the occurrence of a prescribing error or pADE. Each pADE was assessed for the type of error, the classification of agent involved, clinical severity of the error, and the time the error occurred. RESULTS: Forty-one patients with a mean age of 3.2 years were included in our study. Of these patients, 19 (46.3%) experienced at least one pADE. We identified 131 pADEs, 129 of which were prescribing errors conferring a rate of 6.8 errors per affected patient or 3.1 errors per patient admitted to the PICU. The most common error found in the study was incorrect dose calculation (48.1%), with intravenous fluids (41.7%), cardiovascular agents (15.0%), and anti-infectives (12.5%) the most common agents involved with an error. The majority of the pADEs in our study were either clinically serious (33.1%) or significant (44.9%) in nature. Nearly one in every four errors required monitoring and/or intervention to prevent harm, and almost all (96.9%) of the prescribing errors were intercepted before reaching the patient. CONCLUSION: This study highlights incorrect dose calculation as the most common prescribing error in a pediatric critical care setting. Intravenous fluids, cardiovascular agents, and anti-infectives were the classes of medication most commonly involved with a pADE. Due to the high-risk nature of medications used and the critical condition of these patients, more than three-quarters of pADEs were considered to be clinically serious or significant in causing patient harm.

Atmospheric deposition and fluxes of organochlorine pesticides and coplanar polychlorinated biphenyls in aquatic environments of Hong Kong, China.

Concentrations of organochlorine (OC) pesticides and coplanar (dioxin-like) polychlorinated biphenyls (PCBs) in bulk deposition were measured at the Mai Po Marshes Nature Reserve (MPMNR) and A Chau, which are both important habitats for waterbirds in Hong Kong. OC pesticides exempted from the Stockholm Convention were present in greater concentrations than those that have been restricted for use in the region. Among the OC pesticides, HCB, sigmaDDTs, and sigmaHCHs were detected. Concentrations of HCB were greater at MPMNR than at A Chau, and this finding suggests short-range transport of this compound, which is different from the other OC pesticides. Several environmental factors including seasonal variations in temperature, particulate matter, and rainfall may influence the flux of OC pesticides. Since sources of HCB often coexist with sources of polychlorinated dibenzo-p-dioxins and dibenzofurans (dioxins and furans), the presence of HCB may be a useful surrogate for monitoring airborne dioxins and for understanding their deposition potential. The contribution of atmospheric deposition to the OC pesticide input to the two study sites was small. Concentrations of most OC pesticides in surface waters were greater than would be predicted based on the inputs from atmospheric deposition and sedimentation. The mass balance calculation suggests a net flux of OC pesticides from bottom sediments to the overlying water column.