Answer To: BME 530 Statistics and Machine Learning Final project This is a group project for a team of 2-3...
Amar Kumar answered on Dec 09 2021
Algorithm 1
Diabetic Retinopathy with CNN
Identify the diabetes stage in a human retina. Kaggle's Diabetic Retinopathy Detection Repository gave the 150 GB picture information.
To see the model in real life, run the cup application under the Deployed Model envelope.
For testing purposes, test pictures have been saved in the Deployed Model envelope.\
For preparing, the ResNet engineering was employed.\
! [Lingering Learning] (https://cdn-pictures 1.medium.com/max/1200/1*ByrVJspW-TefwlH7OLxNkg.png)
Convolutional Neural Networks, regularly known as ConvNets, are a sort of profound discovering that is most normally used to dissect picture assortments. \
ConvNets have an assortment of layers that are regularly utilized.
* Convolution layer - This layer performs convolution on a picture with indicated step and cushioning.
* Pooling layer - This layer is utilized to decrease the dimensionality of component maps by first building up a veil and an activity to be done, then, at that point, moving the cover over the entire picture as indicated by the step. In this layer, no loads are learned.
* Completely Connected layer - Traditional neural layers found at the neural organization's terminal stem. Because of the surprising number of boundaries it needs, it is just utilized rarely nowadays.
* Dropout layer - This layer is utilized to lessen over-fitting. During the preparation, it turns down specific neurons at arbitrary.
* Bunch Normalization - This lessens estimation time by normalizing the result esteems. It additionally has a regularization impact.
#### I composed a post [here](https://medium.com/@s.ganjoo96/diabetic-retinopathy-recognition with-resnet50-b621514bd22b) on distinguishing diabetic retinopathy.
## Arrangement
### Clone this store and download the information
* Save this archive to your PC as a clone.
* Utilize the album Diabetic-Retinopathy-Detection-with-CNN to get into the envelope.
*Acquire the information records from Kaggle and save them in this area.
### Setting up the essentials
* requirements.txt for pip introduce
## Appropriateness
* In the Model script.ipynb document, run every cell.
* Save your model in the Deployed Model organizer's subfolder.
* Open the app.py document and run it.
* After the server has completed the process of stacking, open an internet browser and type localhost:5000 into the location bar.
#### You are good to go to go.
## Result
Due to the restricted monetary assets, this model was prepared on the cloud for **only**2 ages and accomplishes a 73 percent exactness.
A Machine Learning Approach for the Diagnosis of Parkinson's Disease via Speech Analysis
Introduction
· Parkinson's sickness, which influences in excess of 10 million individuals internationally, is the second most normal neurological disease after Alzheimer's. Parkinson's sickness is portrayed by a decrease in engine and intellectual capacities.
· There is nobody test that can be utilized to analyze a condition. Specialists should rather do a careful clinical assessment of the patient's clinical history.
· Lamentably, this technique for analysis is very insufficient. As indicated by the National Institute of Neurological Disorders, early analysis (manifestations for under 5 years) is just 53% exact. This isn't obviously superior to taking a blind leap of faith, however early location is fundamental for powerful treatment.
· Due to these difficulties, I use a dataset of various discourse qualities (a non-obtrusive yet unmistakable procedure) from the University of Oxford to concentrate on an AI methodology to appropriately analyze Parkinson's sickness.
· What are the advantages of discourse highlights? Since basically every Parkinson's patient has critical vocal decay (powerlessness to create delayed phonations, quake, and roughness), it's a good idea to use voice to recognize the condition. Voice investigation additionally enjoys the benefit of being non-intrusive, minimal expense, and easy to extricate clinically.
Background
Parkinson's Disease
· Parkinson's sickness is a dynamic neurological illness brought about by the deficiency of...