Measuring Vertical Jump Height With Artificial Intelligence Through a Cell Phone: A Validity and Reliability Report

Area of Science:

• Sports Science
• Biomechanics
• Artificial Intelligence in Sports

Background:

• Accurate measurement of vertical jump height is crucial for assessing athletic performance and training progress.
• Traditional methods using force platforms can be expensive and less accessible.
• Mobile applications offer a potential for widespread, convenient performance monitoring.

Purpose of the Study:

• To evaluate the reliability and validity of an artificial intelligence (AI)-driven mobile application (My Jump 2) for estimating vertical jump height.
• To compare the AI model's performance against a gold-standard Force Platform (FP).

Main Methods:

• A cross-sectional study involving 88 athletes (33 female, 55 male) performing 264 countermovement jumps.
• Simultaneous measurement of jump height using the Force Platform (FP) and the My Jump 2 app.
• My Jump 2 utilized a flight-time method with AI for automated jump detection, compared to FP's impulse-momentum method.

Main Results:

• The AI model demonstrated high reliability (ICC = 0.980, CV = 4.12) and the FP also showed high reliability (ICC = 0.990, CV = 2.92).
• Strong agreement was found between the AI model and FP for validity (ICC = 0.973).
• Bland-Altman analysis and regression confirmed no significant systematic or proportional bias.

Conclusions:

• The AI-driven My Jump 2 app is a highly reliable and valid tool for estimating vertical jump height.
• Strength and conditioning professionals can utilize this AI-based mobile app as a practical and efficient alternative for jump height assessment.