# Force-sensing drill-through detection and automatic stopping in a porcine distal humerus model simulating pediatric supracondylar pinning

## TL;DR

A robotic system with force sensing can significantly reduce overdrilling during K-wire placement in a model of children's elbow fractures, improving safety.

## Contribution

A force-sensing drill-through stopping system is introduced and tested for reducing overdrilling in pediatric orthopedic procedures.

## Key findings

- The robotic system reduced overdrill depth by 86.8% compared to manual drilling.
- Higher feed rates modestly increased overdrill depth, while spindle speed had no significant effect.
- Overdrill depth was limited to approximately 1 mm with the robotic system.

## Abstract

Supracondylar humeral fractures are the most common elbow fractures in children and are often treated with closed reduction and percutaneous Kirschner wire (K-wire) fixation. After far-cortex breach, delayed stopping can cause overdrilling and jeopardize adjacent neurovascular structures. We evaluated a force-sensing drill-through stopping system in a porcine humerus model and assessed the effects of feed rate and spindle speed on overdrill depth.

Porcine distal humeri were drilled with a 2.0-mm K-wire introduced through the lateral condyle at ∼60° to the humeral longitudinal axis. A six-axis force/torque sensor measured axial force; a transient force drop triggered an automatic stop command. Overdrill depth was the distance the K-wire tip advanced beyond the outer surface of the far cortex. Parameter tests compared feed rates (0.5/1.0/1.5 mm·s−¹ at 1,200 r min−¹) and spindle speeds (900/1,200/1,500 r min−¹ at 1.0 mm s−¹). Robotic vs. manual drilling was evaluated in paired tests at adjacent, non-interfering sites; manual drilling was performed by a senior pediatric orthopedic surgeon using tactile feedback. Statistical analysis used repeated-measures one-way ANOVA with Geisser–Greenhouse correction and Tukey post hoc tests, and paired t-tests (α = 0.05).

Feed rate significantly affected overdrill depth (Geisser–Greenhouse corrected, p = 0.016); 1.5 mm·s−¹ produced greater overdrill depth than 0.5 mm·s−¹ (Δ = 0.252 mm, adjusted p = 0.0456). Spindle speed had no significant effect (p = 0.900). In paired comparisons, the robotic system reduced overdrilling from 6.60 ± 1.53 mm (manual) to 0.87 ± 0.12 mm (robotic) (mean paired difference 5.73 mm, 95% CI 4.67–6.80; p < 0.0001), an 86.8% reduction.

The force-sensing drill-through stopping system limited overdrill depth to approximately 1 mm in a porcine humerus model. Within the tested range of 900–1,500 r·min−¹ and 0.5–1.5 mm·s−¹, higher feed rates produced a modest increase in overdrilling whereas spindle speed had no significant effect. Compared with manual drilling, the system substantially reduced overdrill depth (≈1 mm vs. 6.6 mm), suggesting potential safety advantages during percutaneous pinning by limiting overdrilling and thereby increasing the safety margin after far-cortex breakthrough. Clinical studies are warranted to determine whether this translates into fewer neurovascular complications.