Skip to main content

A New Problem: Heart Disease

Next, we'll see if we can tackle a problem that may have real-world implications for deep learning. While I jest about the awesome power of neural networks for classifying flowers, there are some cool ways to show that the long arm of deep learning extends beyond the casual gardener's domain.

Specifically, I'd like to show that deep learning can be used in a healthcare setting. There are plenty of ways that I might be able to show this, but perhaps the message would be most effective if I could tackle something relevant. So, we'll be tackling a prediction task related to heart disease, which, according to this image, is rather important:



Heart disease affects most people whether it's directly or through the affliction of a loved one. It is the cause of death for 25% of Americans and, if your household is as pork-chop-laden as mine, probably has affected you as well.

So, what then can we do to show that deep learning is relevant to heart disease? Well, it turns out that there is a set of patient data publicly available that will allow us to build a classifier: (http://archive.ics.uci.edu/ml/datasets/heart+Disease).

According to this website, the dataset contains 75 attributes (though they only use 14 attributes) and contains 303 samples. This isn't an extremely large dataset, but we'll see if we can come up with a neural network to use this data. Note that extremely complex classification tasks, such as image recognition, requires an enormity of data that would surpass the volume given in this dataset.

So let's go ahead and see what kind of attributes we're working with.

The website states that the following features are given for each case:
1. Age: age of the patient in years.
2. Sex: 1 if male; 0 if female.
3. Cp: chest pain type (1. typical angina, 2. atypical angina, 3. non-anginal pain, 4. asymptomatic)
4. Trestbps: resting bp in mmHg
5. Chol: serum cholesterol in mg/dl
6. fbs: bit if fasting blood sugar > 120mg/dl (1 = true, 0 = false)
7. restecg: resting electrocardiographic results (0: normal, 1: ST-T wave abnormality)
8. thalach: maximum heart rate achieved
9. exang: exercise induced angina (1: yes, 0: no)
10. oldpeak: ST depression induced by exercise relative to rest
11. slope: the slope of the peak exercise ST segment (1: upsloping, 2: flat, 3: downsloping)
12. ca: number of major vessels (0-3) colored by flouroscopy
13. thal: (3: normal, 6: fixed defect, 7: reversible defect)
14. num (predicted attribute): diagnosis of heart disease (0: no, 1: yes)

So it looks like we're given some demographic information for each patient (i.e. the age and sex of the patient). We're also given some acute information about a patient, such as whether there's angina (chest pain), what the resting blood pressure is, cholesterol levels, etc. All these pieces of information are important in determining the likelihood that a patient develops heart failure. In addition to these measurements, we also have a boolean value that describes whether the patient has heart failure. That's good.

Given these measurements, we can conceive of a function that we might approximate with a neural network. We're given these measurements and whether the patient develops heart failure, so we can see that we can use those given measurements to build a classifier that predicts the onset of heart failure. For this network, the measurements and demographic data would be used as input to the network, and the presence of heart failure would be used as an output.

That's a good problem definition. Let's continue this problem by pre-processing the data in R in the next post.


Comments

Post a Comment

Popular posts from this blog

Classifying Flowers Part 1: Data

Everybody loves plants. So we'll continue our descent by building a neural network that can do something with plants. This construct will be able to take characteristics that describe a plant and, if we're good enough, hopefully the neural network will be able to tell us with accuracy what type of plant it is. To do this, we'll use the Iris flowers dataset, which contains 150 samples of 3 different species (50 samples each). The species in this dataset are Iris setosa, Iris versicolor, and Iris virginica. The dataset can be found at the following links. Data:  https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data ReadME:  https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.names Here's a screenshot of the data we'll be working with: Initially, this data doesn't look too exciting and if it at all confuses you, you're perfectly normal. But we can figure some immediate information just from looking at the format ...
Post a Comment
Read more

Classifying Flowers Part 2: Training and Evaluation

In the previous post, we fiddled with data and set out for an ambitious task of classifying plants. We ended that post after having downloaded, prepared, and partitioned our data, resulting two non-overlapping sets of data: a test set and a training set. We'll continue now by building, training, and evaluating a neural network to classify these flowers. First, we'll need to separate the values that we'll use as input to the network from the values we're intending to use as the output from the network. In other words, we need to place all the sepal and petal widths and lengths into a bag that the network can read, and all the species corresponding to the individual sets of sepal and petal widths and lengths into another bag. That way, when the network trains, it only takes into account the 4 inputs. When it produces an output, we can use the other bag (the one with species inside) to see if the network has predicted the species correctly. Before doing that, we'...
Post a Comment
Read more

Evaluating our Heart Disease Classifier

Let's continue our descent by evaluating how good our heart disease classifier is. We can do this by generating predictions on the test set and see how the predictions compare to the test set's ground truth labels. That can be done with the following lines of code: At the bottom we can see that the classifier predicted 24 true negatives, 9 false positives, 8 false negatives, and 19 true positives. That's pretty okay. There is obviously some inaccuracy in the predictions, but let's calculate the accuracy anyways. (24+19)/(24+19+8+9)=71.6. So the test accuracy was 71.6, while, if you recall from the last post, the training accuracy was nearing 90%. This disparity between training and testing accuracy is a result of overfitting. Essentially, 50,000 training loops was too much training for this little of data. The resulting network overfit to the noise inherent in the training data and, as a result, failed to generalize as well on the test set. Therefore, the testing...
Post a Comment
Read more