Validity of the iPhone M7 motion co-processor as a pedometer for able-bodied ambulation

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Physical Activity Monitoring

Background:

• Physical activity is crucial for disease prevention.
• Smartphones can monitor physical activity through step counting.
• Hardware-software variability challenges accurate step count estimation.

Purpose of the Study:

• To validate the Apple iPhone M7 motion co-processor for step count estimation.
• To assess the M7's accuracy across different walking speeds.
• To compare M7 step counts with a StepWatch device and manual counts.

Main Methods:

• Twenty participants walked at slow, customary, and fast speeds on level ground.
• Step counts were simultaneously recorded using an iPhone with a custom app, a StepWatch device, and manual counting.
• Data analysis involved Pearson correlation and Bland-Altman analyses.

Main Results:

• Strong correlations between iPhone and manual/StepWatch counts were observed for customary and fast walking speeds.
• Weak correlation was found for slow walking speeds.
• Mean absolute errors were 21% (slow), 8% (customary), and 4% (fast), indicating higher accuracy at faster speeds.

Conclusions:

• The M7 motion co-processor accurately records step counts during customary and fast walking.
• Inaccuracies at slow speeds have implications for specific patient populations.
• The iPhone may serve as a viable alternative to the StepWatch for faster walking speeds only.