User Profile

Hello, I am working on a clinical-grade application with access to the CMMovementDisorderManager API to track Parkinson's symptoms. Sometimes, these two method calls produce an NSError in the CMErrorDomain with code 103, which appears to be CMErrorUnknown: [CMMovementDisorderManager queryDyskineticSymptomFromDate:toDate:withHandler:] [CMMovementDisorderManager queryTremorFromDate:toDate:withHandler:] This is difficult for us to track down, because the error code is nondescript, and the user info is empty. We really don't have any clues as to what could be causing this behavior. I'm hoping a CoreMotion framework engineer with visibility into the code may be able to take a look and let us know what conditions can cause this result. Thanks! Dan

Hello, My company is interested in querying the new mobility metrics introduced in iOS 14 / watchOS 7. I have been having trouble collecting samples for some of these new data streams on my devices. I have an iPhone 7 and an Apple Watch Series 4. They are both upgraded to the latest version of the iOS 14 beta watchOS 7 beta. The data streams that have worked are the VO2max and the six minute walk test. The VO2max seems to require a workout session. The six minute walk test data first showed up after wearing the watch consistently for a week, as described in the WWDC video. However, the other data streams (walking speed, step length, walking asymmetry, double support time, stair speed up, and stair speed down) are not automatically generating samples. I was able to manually enter samples in the Health app, but we are interested in the automatically collected data. The walking speed, step length, walking asymmetry, and double support time sections in the Health app says that you have to hold your phone in your pocket and walk over flat ground to record this data. I have gone for several walks (~20 minute / 1 mile) up and down my street with my watch on my wrist and my iPhone in my pocket. I also enabled an outdoor walk workout to make it extra explicit that I was walking. Still, none of these four data streams automatically collected any data. So I am wondering how flat does the ground have to be for this to work? My street is certainly not 100% level; it has small and gentle slopes over it’s length. Do I need to find a larger indoor course or perhaps an outdoor track where the ground is really perfectly level? If so, how much distance is required for the samples to be computed? Similarly, the stair speed up and stair speed down sections in the Health app says the Apple Watch will record this data when you go up or down a flight of stairs that is at least 10 feet tall. I have gone up and down flights of stairs in my house over ten times to attempt to collect this data without success. I measured the vertical distance with a tape measure, and found that my staircases are only 9 feet tall from the starting floor to the ending floor. I tried to compensate for this by going up two cases of stairs in one try (eg. From the basement to the second floor), but that didn’t work either. I assume that the stretch of level ground from the top of the basement stairs and around to the bottom of the main stairs was enough for the watch to break up the stair climbs into two separate 9 foot instances instead of a single 18 foot instance. Can you confirm that this is the case and let me know if I should expect this functionality to work if I find a building with bigger stairs and tighter turns? Interestingly, under the Activity section of the Health app, there are some metrics that seem related to these mobility metrics. For example, the Health app is recording my walking distance and step count, from which I could imagine computing the step length as a simple division (distance per step). Since the walking samples are timestamped, it may also be possible to compute the walking speed as the distance over duration. Finally, the number of stair flights climbed is also timestamped. So again, the stair up/down speed might be computed as vertical distance per second. Perhaps the estimation algorithms are more sophisticated, and so the computations do not follow so easily from these other data points, but I thought I would mention this observation in case it indicates a problem with my data collection approach. Thanks for any clarity you can provide on the details of the requirements for the data collection on these streams! Cheers, Dan Jackson Amato Senior Engineer, Digital Artefacts