PurposeThe purpose of this study was to determine whether computed tomography (CT)-based machine learning of radiomics features could help distinguish autoimmune pancreatitis (AIP) from pancreatic ductal adenocarcinoma (PDAC).Materials and MethodsEighty-nine patients with AIP (65 men, 24 women; mean age, 59.7 ± 13.9 [SD] years; range: 21–83 years) and 93 patients with PDAC (68 men, 25 women; mean age, 60.1 ± 12.3 [SD] years; range: 36–86 years) were retrospectively included. All patients had dedicated dual-phase pancreatic protocol CT between 2004 and 2018. Thin-slice images (0.75/0.5 mm thickness/increment) were compared with thick-slices images (3 or 5 mm thickness/increment). Pancreatic regions involved by PDAC or AIP (areas of enlargement, altered enhancement, effacement of pancreatic duct) as well as uninvolved parenchyma were segmented as three-dimensional volumes. Four hundred and thirty-one radiomics features were extracted and a random forest was used to distinguish AIP from PDAC. CT data of 60 AIP and 60 PDAC patients were used for training and those of 29 AIP and 33 PDAC independent patients were used for testing.ResultsThe pancreas was diffusely involved in 37 (37/89; 41.6%) patients with AIP and not diffusely in 52 (52/89; 58.4%) patients. Using machine learning, 95.2% (59/62; 95% confidence interval [CI]: 89.8–100%), 83.9% (52:67; 95% CI: 74.7–93.0%) and 77.4% (48/62; 95% CI: 67.0–87.8%) of the 62 test patients were correctly classified as either having PDAC or AIP with thin-slice venous phase, thin-slice arterial phase, and thick-slice venous phase CT, respectively. Three of the 29 patients with AIP (3/29; 10.3%) were incorrectly classified as having PDAC but all 33 patients with PDAC (33/33; 100%) were correctly classified with thin-slice venous phase with 89.7% sensitivity (26/29; 95% CI: 78.6–100%) and 100% specificity (33/33; 95% CI: 93–100%) for the diagnosis of AIP, 95.2% accuracy (59/62; 95% CI: 89.8–100%) and area under the curve of 0.975 (95% CI: 0.936–1.0).ConclusionsRadiomic features help differentiate AIP from PDAC with an overall accuracy of 95.2%. 相似文献
BackgroundThe measurement of health-related quality of life (HRQoL) provides information about the perceived burden of the health condition and treatments from a lived experience. The Brisbane Burn Scar Impact Profile (caregiver report for young children, BBSIP0–8), developed in 2013, is a proxy-report measure of burn scar-specific HRQoL. The aim of this study was to report its psychometric properties in line with an evaluative purpose.MethodsCaregivers of children up to 8 years of age at risk of burn scarring were recruited into a prospective, longitudinal cohort study. Caregivers completed the BBSIP0–8, Pediatric Quality of Life Inventory and Patient Observer Scar Assessment Scale at baseline (approximately ≥85% of the total body surface area re-epithelialised), 1–2 weeks after baseline and 1-month after baseline. Psychometric properties measured included internal consistency, test–retest reliability, validity and responsiveness.ResultsEighty-six caregivers of mostly male children (55%), of a median age (IQR) of 1 year, 10 months (2 years, 1 month) and total body surface area burn of 1.5% (3.0%) were recruited. Over one third of participants were grafted and 15% had contractures or skin tightness at baseline. Internal consistency of ten item groups ranged from 0.73 to 0.96. Hypothesised correlations of changes in the BBSIP0–8 items with changes in criterion measures supported longitudinal validity (ρ ranging from ?0.73 to 0.68). The majority of item groups had acceptable reproducibility (ICC = 0.65–0.83). The responsiveness of five item groups was supported (AUC = 0.71–0.90).ConclusionThe psychometric properties tested support the use of the BBSIP0–8 as an evaluative measure of burn scar-related health-related quality of life for children aged below eight years in the early post-acute period of rehabilitation. Further investigation at longer time period after burn injury is indicated. 相似文献
Objective: Longitudinal data on cardiometabolic effects of egg intake during adolescence are lacking. The current analyses aim to evaluate the impact of usual adolescent egg consumption on lipid levels, fasting glucose, and insulin resistance during late adolescence (age 17–20?years).
Methods: Data from 1392 girls, aged 9 to 10 at baseline and followed for 10?years, in the National Heart, Lung, and Blood Institute’s National Growth and Health Study were used to examine the association between usual egg intake alone and in combination with other healthy lifestyle factors and late adolescent lipid levels, fasting glucose, and insulin resistance, measured as homeostasis model assessment of insulin resistance (HOMA-IR). Diet was assessed using 3-day food records during eight examination cycles. Girls were classified according to usual weekly egg intake, ages 9–17?years:?<1 egg/wk (n?=?361), 1 to <3 eggs/wk (n?=?703), and ≥3 eggs/wk (n?=?328). Analysis of covariance modeling was used to control for confounding by other behavioral and biological risk factors.
Results: Girls with low, moderate, and high egg intakes had adjusted low-density lipoprotein cholesterol levels of 99.7, 98.8, and 95.5 mg/dL, respectively (p?=?0.0778). In combination with higher intakes of fiber, dairy, or fruits and vegetables, these beneficial effects were stronger and statistically significant. There was no evidence that ≥3 eggs/wk had an adverse effect on lipids, glucose, or HOMA-IR. More active girls who consumed ≥3 eggs/wk had the lowest levels of insulin resistance.
Conclusion: These results suggest that eggs may be included as part of a healthy adolescent diet without adverse effects on glucose, lipid levels, or insulin resistance. 相似文献
Long-term follow-up of patient-reported outcome measures (PROM) is essential in both modern spinal care and research. Lack of time and staff are commonly reported barriers to implementing long-term follow-up of PROM. Automated and digital follow-up systems for PROM collection are seeing widespread use, yet their validity and comparative effectiveness have never been evaluated.
Purpose
The present study aimed to assess the validity of digital follow-up systems in comparison with the conventional paper-based follow-up (PB-FU).
Study Design
This is a retrospective analysis of prospectively collected double follow-up data.
Patient Sample
Patients who underwent lumbar spinal fusion for spondylolisthesis or degenerative disc disease between 2013 and 2016 were included in the study.
Outcome Measures
The study determined the Oswestry Disability Index (ODI) and Numeric Rating Scale (NRS) for back and leg pain severity at baseline, 6 weeks, 12 months, and 24 months.
Materials and Methods
After lumbar spinal fusion surgery, a double follow-up of PROM was carried out by conventional PB-FU during clinical visits, while simultaneously completing an automatically dispatched digital follow-up questionnaire. As the primary end point, we assessed the intraindividual discrepancy in PROM between PB-FU and automated digital follow-up (AD-FU).
Results
Forty patients completed all parts of the dual follow-up trajectory and were analyzed. We detected no discrepancy in ODI or NRS for back and leg pain severity at any of the baseline, 6-week, 12-month, or 24 month follow-ups (all p>.05). This was confirmed in a sensitivity analysis.
Conclusions
In an analysis of dual paper-based and digital follow-up after lumbar fusion surgery, patients report highly similar values using either method of follow-up. It appears that AD-FU without incentives produces lower response rates. To reassess the validity of these systems for data collection in spinal patient care, a prospective validation with higher statistical power is warranted. 相似文献