A machine learning approach for predicting radiation-induced hypothyroidism in patients with nasopharyngeal carcinoma undergoing tomotherapy.

The purpose of this study was to establish an integrated predictive model that combines clinical features, DVH, radiomics, and dosiomics features to predict RIHT in patients receiving tomotherapy for nasopharyngeal carcinoma. Data from 219 patients with nasopharyngeal carcinoma were randomly divided into a training cohort (n = 175) and a test cohort (n = 44) in an 8:2 ratio. RIHT is defined as serum thyroid-stimulating hormone (TSH) greater than 5.6 µU/mL, with or without a decrease in free thyroxine (FT4). Clinical features, 27 DVH features, 107 radiomics features and 107 dosiomics features were extracted for each case and included in the model construction. The least absolute shrinkage and selection operator (LASSO) regression method was used to select the most relevant features. The eXtreme Gradient Boosting (XGBoost) was then employed to train separate models using the selected features from clinical, DVH, radiomics and dosiomics data. Finally, a combined model incorporating all features was developed. The models were evaluated using receiver operating characteristic (ROC) curves and decision curve analysis. In the test cohort, the area under the receiver operating characteristic curve (AUC) for the clinical, DVH, radiomics, dosiomics and combined models were 0.798 (95% confidence interval [CI], 0.656-0.941), 0.673 (0.512-0.834), 0.714 (0.555-0.873), 0.698 (0.530-0.848) and 0.842 (0.724-0.960), respectively. The combined model exhibited higher AUC values compared to other models. The decision curve analysis demonstrated that the combined model had superior clinical utility within the threshold probability range of 1% to 79% when compared to the other models. This study has successfully developed a predictive model that combines multiple features. The performance of the combined model is superior to that of single-feature models, allowing for early prediction of RIHT in patients with nasopharyngeal carcinoma after tomotherapy.

PYHIN1 correlates with CD8+ T cells infiltration and confers good patient survival in oral cancer.

BACKGROUND/PURPOSE: Immunotherapy has become a research hotspot and is used for head and neck cancer treatment. This research aims to explore the prognostic value of PYHIN1 in oral cancer and the relationship between PYHIN1 and cancer immunity. MATERIALS AND METHODS: The expression of PYHIN1 in clinical specimens was evaluated by bioinformatics analyses and immunohistochemistry. RESULTS: Gene ontology term enrichment analyses and gene set enrichment analyses showed the involvement of PYHIN1 in the modulation of adaptive immunity-associated signaling according to The Cancer Genome Atlas database and Gene Expression Omnibus dataset. Interestingly, the correlation analyses in The Cancer Genome Atlas database revealed a positive correlation between PYHIN1 expression and activated CD8+ T cells infiltration and a negative correlation between PYHIN1 expression and tumor purity. Moreover, activated CD8+ T cells infiltration predicted good patient survival and was negatively correlated with tumor purity. Importantly, PYHIN1 expression was negatively correlated with the pathological stage and was positively associated with a good prognosis in patients with oral cancer. The data obtained from the Gene Expression Omnibus dataset and immunohistochemistry confirmed the positive association between PYHIN1 and CD8+ T cells infiltration in oral cancer tissues. CONCLUSION: We conclude that PYHIN1 is an indicator of cancer immunity, and is an independent prognostic factor that may be an alternative target for oral cancer treatment.

Correction: A knowledge, attitude, and practice survey on mediation among clinicians in a tertiary-care hospital in Singapore.

[This corrects the article DOI: 10.1371/journal.pone.0199885.].

A knowledge, attitude, and practice survey on mediation among clinicians in a tertiary-care hospital in Singapore.

Healthcare delivery is a highly complex, deeply personal and costly endeavour that involves multiple specialties and services. There is an imbalance in knowledge between the healthcare provider and consumer that may contribute to doubts and uncertainty over treatment and outcomes. It is unsurprising that conflict and dispute can develop between healthcare providers and patients and their next-of-kin. The use of mediation in the healthcare setting has recently been promoted in many developed countries, including Singapore. We administered a detailed 32-item survey in a large tertiary-care teaching hospital to improve our understanding of the knowledge, attitude and practice of dispute resolution among clinicians to pave the way for better strategies to improve the adoption of mediation in healthcare setting. Ninety-seven respondents had an average of 62% (SD: 12%) knowledge score. The most common misconceptions held by the respondents about mediation were: (1) mediation was about fact-finding, (2) mediation is limited to only certain types of dispute, (3) mediation proceeds by both parties giving their account of the dispute, then a third party decides a settlement, (4) the average time it takes to resolve a dispute through mediation, (5) the cost of mediation, (5) the venue of mediation, (6) the person determining the outcome of mediation, (7) confidentiality of mediation. In general, the respondents were positive about the use of mediation as a dispute resolution tool. When asked to indicate the relative importance of different outcomes of dispute resolution, financial compensation and waiver of hospital bill attracted mixed responses while understanding facts of dispute, assurance that the same error would not recur, and offering corrective treatment were rated as being important. By contrast, seeking an apology from the complainant was considered neutral to somewhat important and the respondents were least concerned with the publicity of the dispute. Direct negotiation with the complainant was considered the most time- and cost-efficient means of resolving a dispute while the opposite was true for litigation. Mediation was considered the approach where the clinicians are most likely to achieve their desired outcome while litigation was considered least likely to produce a favourable outcome. Approximately half of the respondents reported having personal experience or known of a colleague who had been involved in a medico-legal dispute. A quarter of these cases were resolved by direct negotiations with the complainant while lawyers, the judge and mediation, resolved approximately 15% each, respectively. The knowledge base of the clinicians in this study about mediation was moderate and probably reflected the general lack of direct experience in the resolution of a dispute or training in mediation. This further corroborated with the general response that the uptake of mediation in the healthcare setting is currently poor in Singapore due to the lack of awareness and perceived lack of avenue among the surveyed clinicians. Any further work to be done with clinicians may be in the direction of (1) increasing general understanding of mediation, (2) increasing awareness of avenues for mediation, and (3) becoming better aware of when to propose mediation.