2026-10-10T19:03:45Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34072204 2026-10-05T17:33:52Z category_1 category_12 category_13 category_734 category_106 portal_5 item_type_3 month_year_10_2026

<p>Leave-one-out cross-validation (LOOCV) performance of limb–socket linear regressions. The predicted R² and error metrics are reported for global and slice-based descriptors. 95% Confidence intervals are reported in square brackets.</p> Vishal Pendse (22970448) Nader Allam (20988585) Calvin C. Ngan (22226838) Elaine Lorette (25316944) Isabelle Morin-Girard (25316947) Jan Andrysek (8945657) Biophysics Cell Biology Genetics Biological Sciences not elsewhere classified Science Policy simple linear regressions simple linear models r &# 178 end clearance play direct geometric relationship anterior &# 8211 explain localized geometry evaluated using leave predict socket geometry limb prosthetic devices limb &# 8211 weaker relationships toward overall socket size sectional area ). socket geometric relationships transradial limb absence sectional descriptors showed proximal &# 8211 indicating limited overfitting transradial socket design limb geometry sectional descriptors limb size socket pairs transradial limbs trimline design surface area aligned using residual limb limb prosthetics across descriptors limb length variable performance suspension strategy strongest results several digital prior research novel spline model performance less able largely predictable interpretable description generally high fourteen participants findings suggest clinician experience based method algorithmic approaches <p>Leave-one-out cross-validation (LOOCV) performance of limb–socket linear regressions. The predicted R² and error metrics are reported for global and slice-based descriptors. 95% Confidence intervals are reported in square brackets.</p> 2026-10-05T17:33:41Z Dataset Dataset 10.1371/journal.pdig.0001767.t003 https://figshare.com/articles/dataset/_p_Leave-one-out_cross-validation_LOOCV_performance_of_limb_socket_linear_regressions_The_predicted_R_sup_2_sup_and_error_metrics_are_reported_for_global_and_slice-based_descriptors_95_Confidence_intervals_are_reported_in_square_brackets_p_/34072204 CC BY 4.0