Radiomics-based decision support tool assists radiologists in small lung nodule classification and improves lung cancer early diagnosis.

BACKGROUND: Methods to improve stratification of small (≤15 mm) lung nodules are needed. We aimed to develop a radiomics model to assist lung cancer diagnosis. METHODS: Patients were retrospectively identified using health records from January 2007 to December 2018. The external test set was obtained from the national LIBRA study and a prospective Lung Cancer Screening programme. Radiomics features were extracted from multi-region CT segmentations using TexLab2.0. LASSO regression generated the 5-feature small nodule radiomics-predictive-vector (SN-RPV). K-means clustering was used to split patients into risk groups according to SN-RPV. Model performance was compared to 6 thoracic radiologists. SN-RPV and radiologist risk groups were combined to generate "Safety-Net" and "Early Diagnosis" decision-support tools. RESULTS: In total, 810 patients with 990 nodules were included. The AUC for malignancy prediction was 0.85 (95% CI: 0.82-0.87), 0.78 (95% CI: 0.70-0.85) and 0.78 (95% CI: 0.59-0.92) for the training, test and external test datasets, respectively. The test set accuracy was 73% (95% CI: 65-81%) and resulted in 66.67% improvements in potentially missed [8/12] or delayed [6/9] cancers, compared to the radiologist with performance closest to the mean of six readers. CONCLUSIONS: SN-RPV may provide net-benefit in terms of earlier cancer diagnosis.

Blood glucose monitoring by insulin-treated pilots of commercial and private aircraft: An analysis of out-of-range values.

AIM: To examine blood glucose measurements recorded as part of the diabetes protocol operated by the UK, Ireland and Austria, which allows commercial airline pilots with insulin-treated diabetes to fly. METHODS: An observational study was conducted in pilots with insulin-treated diabetes, granted medical certification to fly commercial or noncommercial aircraft, who recorded pre-flight and hourly in-flight blood glucose measurements. These values were correlated to a traffic light system (green 5.0 to 15.0 mmol/L; amber 4.0 to 4.9 mmol/L and 15.1 to 20.0 mmol/L; and red <4.0 mmol/L or >20.0 mmol/L) and studied for trends in glucose concentrations, time course within flight and any consequences. Pilot demographics were also analysed. RESULTS: Forty-four pilots (90%) recorded one or more blood glucose value outside the green range during the 7 years of the study. Pilot age, diabetes type and duration, and follow-up period were comparable among subgroups, and mean glycated haemoglobin did not differ before and after certification in a way which would indicate poorer glycaemic control in any subgroup. A total of 892 blood glucose values (2.31%) were outside the green range, with half reported in-flight at various time intervals. There were 48 (0.12%) low red range values recorded, 14 (0.04%) of which occurred in-flight; all but four were restored to within the green range by the time of the next measurement. Appropriate corrective action was taken for all out-of-range values, with no reports of pilot incapacitation from any cause. CONCLUSIONS: The traffic light system appears effective in identifying and reducing the frequency and severity of out-of-range values.

An Evaluation of the Safety of Pilots With Insulin-Treated Diabetes in Europe Flying Commercial and Noncommercial Aircraft.

OBJECTIVE: The risk of hypoglycemia in people with insulin-treated diabetes has debarred them from certain "safety-critical" occupations, including flying commercial aircraft. This report evaluates the effectiveness of a protocol enabling a large cohort of insulin-treated pilots to fly commercially. RESEARCH DESIGN AND METHODS: This was an observational study of pilots with insulin-treated diabetes who were granted medical certification to fly commercial and noncommercial aircraft. Clinical details, pre- and in-flight (hourly and 30 min before landing) blood glucose values were correlated against the protocol-specified ranges: green (5-15 mmol/L), amber (low, 4-4.9 mmol/L; high, 15.1-20 mmol/L), and red (low, <4 mmol/L; high, >20 mmol/L). RESULTS: A total of 49 pilots with type 1 (84%) or type 2 (16%) diabetes who had been issued class 1 or class 2 certificates were studied. Median diabetes duration was 10.9 years. Mean HbA1c was 7.2% (55.0 mmol/mol) before certification and 7.2% (55.1 mmol/mol) after certification (P = 0.97). Blood glucose values (n = 38,621) were recorded during 22,078 flying hours. Overall, 97.69% of measurements were within the green range, 1.42% within the low amber range, and 0.75% within the high amber range. Only 0.12% of readings were within the low red range and 0.02% within the high red range. Out-of-range readings declined from 5.7% in 2013 to 1.2% in 2019. No episodes of pilot incapacitation occurred, and glycemic control did not deteriorate. CONCLUSIONS: The protocol is practical to implement, and no events compromising safety were reported. This study represents what is, to our knowledge, the most extensive data set from people with insulin-treated diabetes working in a "safety-critical" occupation, which may be relevant when estimating risk in other safety-critical occupations.