Back to Reference Library
veterinary
farriery
2023
Cohort Study

Detection of foaling using a tail-attached device with a thermistor and tri-axial accelerometer in pregnant mares.

Authors: Aoki Takahiro, Shibata Makoto, Violin Guilherme, Higaki Shogo, Yoshioka Koji

Journal: PloS one

Summary

# Editorial Summary Attending a mare during labour requires reliable early warning of imminent foaling, yet current detection methods remain inconsistent; Aoki and colleagues developed a multimodal sensor affixed to the ventral tail base that monitors surface temperature, roll angle (postural changes), and tail acceleration to identify labour onset. Data collected every three minutes from 17 pregnant mares demonstrated exceptional accuracy in distinguishing standing from recumbent positions (Cohen's kappa = 0.99), with the final prepartum hour characterised by significantly lower surface temperatures (<35.5°C), prolonged recumbency, and tail-raising behaviour compared to the preceding 2–120 hours (P < 0.01). Single indicators achieved 100% sensitivity but proved unreliable in isolation (precision ranging from 2.8–13.1%); however, combining lower surface temperature with either lying-down behaviour or tail-raising substantially improved precision to 32.1–56.7%, whilst the three-parameter combination (temperature + recumbency + tail elevation) achieved 100% precision. For practitioners, this technology offers a practical means of detecting labour within the critical final hour before delivery, potentially enabling timely intervention to manage complications, though further validation across larger mare populations and diverse management systems would strengthen clinical applicability.

Read the full abstract on PubMed

Practical Takeaways

  • A wearable tail-base sensor device can reliably alert you to imminent foaling when multiple parameters (temperature drop, recumbency, tail-raising) are present together, enabling timely attendance at delivery.
  • Single sensor indicators alone produce false alarms; combining three parameters eliminates false positives for practical farm use.
  • This technology could reduce complications from unattended foaling by providing advance warning within the final hour before labor onset.

Key Findings

  • A tail-attached sensor device with thermistor and accelerometer accurately distinguished standing vs. recumbent postures (Cohen's kappa 0.99) in pregnant mares.
  • Combined indicators of lower surface temperature (<35.5°C) + lying down + tail-raising achieved 100% precision for foaling detection within one hour before parturition.
  • Individual sensor parameters showed 100% sensitivity but poor precision (2.8-13.1%); combining parameters LST+LD+TR improved detection to 100% precision.
  • Hourly sensor data patterns differed significantly between the last hour prepartum versus 2-120 hours before parturition (P<0.01).

Conditions Studied

pregnancy monitoringfoaling detectionlabor onset