Fitness watch configured with micro AI
A system and method for biologically monitoring the fitness of an athlete, and providing a warning when an overtraining condition is determined in order to reduce injury. Through implementation of an efficient system architecture, micro-artificial intelligence use is practical for mobile situations where internet coverage is deficient or non-existent.
1. A fitness watch, comprising:
a heart rate sensor configured to provide heart rate data of the user;
a bifurcated memory structured to reduce a quantity of data subject to an artificial intelligence analysis, said bifurcated memory including:
a latent memory configured to retain latent data including demographic; and historical, non-current exercise data, said latent memory being periodically updatable defined by when the fitness watch is within an internet coverage area;
a current memory configured to retain current data that is updatable as the user is exercising, irrespective of internet coverage, the current data including heart rate data provided by said heart rate sensor, said latent memory and said current memory being differentially updatable relative to one another, depending upon presence of the fitness watch in the internet coverage area; and
a microprocessor including a neural network, said microprocessor being configured to determine an existence of an overtraining condition based on an output of the neural network utilizing only updatable and periodically updatable data in said bifurcated memory to enable operation of the neural network in a mobile environment, said microprocessor being configured to provide an alert to the user after determining that the overtraining condition exists according to the output of the neural network.
2. The fitness watch of claim 1 , wherein the determination is based on active heart rate of the user while the user was exercising.
3. The fitness watch of claim 1 , wherein the determination is based on a combination of active heart rate and at least one of a training intensity level and training duration.
4. The fitness watch of claim 1 , wherein said microprocessor is configured to compare said current data with said latent data to determine the existence of the overtraining condition.
5. The fitness watch of claim 1 , wherein said neural network utilizes a Bayesian classifier to generate the output.
6. The fitness watch of claim 1 , wherein said microprocessor is configured to determine the existence of an overtraining condition as the user is exercising.
7. The fitness watch of claim 1 , wherein said microprocessor is configured to determine the existence of the overtraining condition based on the output of the neural network while the user is exercising.
8. The fitness watch of claim 1 , wherein the determination of the existence of the overtraining condition is triggered by a weighted trigger condition.
9. The fitness watch of claim 8 , wherein a primary trigger is heart rate, and a secondary trigger is sleep duration.
10. The fitness watch of claim 1 , wherein said microprocessor is configured to provide the alert to the user while the user is exercising.
11. The fitness watch of claim 1 , wherein the output of the neural network is based on at least two classifiers, one of the classifiers being a Bayesian classifier.
12. The fitness watch of claim 1 , wherein said microprocessor is configured to generate a fitness condition based at least on a combination of active heart rate and pace.
13. The fitness watch of claim 1 , wherein said microprocessor is configured to generate a fitness condition based on sleep time.
14. The fitness watch of claim 1 , wherein the latent data includes geographic data.
15. The fitness watch of claim 1 , further comprising a movement sensor.
16. The fitness watch of claim 1 , wherein said microprocessor is configured to provide an alert cautioning the user to alter training frequency.
17. The fitness watch of claim 1 , wherein said microprocessor is configured to provide an alert warning the user to alter a duration of exercise.