# Processing with AI ## Exploration : IA & Ethics Name: > Jules Ehrbar Problematic: > Sports Medicine Small description of the project: > Use AI to do a pre-diagnosis of the corpulence of the patient with a picture of his/her body >[TOC] ## Design brief ### Biais If we don't source our dataset with enough rigor, the following biais might appear: >1. gender and ethnicity >2. shooting style >3. age We will ensure that our model is not biaised by: >1. Sourcing our data from diffenrent genders et ethnicities >2. Making sure our data take into account different age ranges >3. Our datas are splitted into differnet shooting style/positions ### Overfitting We will make sure our model does not overfit by > Checking the accuracy of our model on a different dataset that the one we trained our model with. ### Misuse >We have to remind ourselves that our application could be misused by insurance companies to gather medical data on their clients and therefor adapt their offer and prices depending on your health situation. ### Data leakage *Choose the most relevant proposition:* >In a catastrophic scenario, where all of our training dataset were stolen or recovered from our model, the risk would be that private medical records would be in the wrong hands. Therefore we need to make sure that our data should be anonymous and without recognizable faces. ### Hacking > If someone found a way to "cheat" our model and make it make any prediction that it want instead of the real one, the risk would be that the standards of "being in good shape" would be changed in its benefit.