Chest X-Ray Medical Diagnosis with Deep Learning - ⑤ Prediction and Evaluation

Chest X-Ray Medical Diagnosis with Deep Learning - ⑤ Prediction and Evaluation 

start

BioinformaticsAndMe

 

 

Chest X-Ray Medical Diagnosis with Deep Learning

     ① Import Packages and Function
     ② Load the Datasets
     ③ Model Development
     ④ Training
    ⑤ Prediction and Evaluation 

 

 

 

 

⑤ Prediction and Evaluation

---

: 테스트셋을 사용하여 모델을 평가

predicted_vals = model.predict(test_generator, steps = len(test_generator))

 

 

 

5.1 ROC Curve and AUROC

---

: ROC(Receiver Operating Characteristic) 곡선에서 AUC(Area Under the Curve)를 계산

: 곡선이 좌상단에 있을수록, 더 많은 면적이 갖고, 모델이 더 잘 수행되고 있음을 의미함

: util.py에서 제공하는 util.get_roc_curve 함수를 사용

auc_rocs = util.get_roc_curve(labels, predicted_vals, test_generator)

 

 

 

5.2 Visualizing Learning with GradCAM

---

: GradCAM 기술을 사용하여 병리학적 상태를 예측하기 위해, 이미지의 중요한 영역을 강조하는 히트맵을 생성

: 먼저 작은 Training 셋을 로드하고, 앞서 AUC 측정값이 가장 높은 4개의 클래스를 살펴볼 것

df = pd.read_csv("drive/MyDrive/nih/train-small.csv")
IMAGE_DIR = "drive/MyDrive/nih/images-small/"

# only show the lables with top 4 AUC
labels_to_show = np.take(labels, np.argsort(auc_rocs)[::-1])[:4]

: 몇 가지 특정 이미지들을 확인

util.compute_gradcam(model, '00008270_015.png', IMAGE_DIR, df, labels, labels_to_show)

# Expected output
Loading original image
Generating gradcam for class Cardiomegaly
Generating gradcam for class Mass
Generating gradcam for class Pneumothorax
Generating gradcam for class Edema

util.compute_gradcam(model, '00011355_002.png', IMAGE_DIR, df, labels, labels_to_show)

util.compute_gradcam(model, '00029855_001.png', IMAGE_DIR, df, labels, labels_to_show)

: P 값은 정답 클래스의 softmax probability를 의미함

 

 

 

 

#Reference

1) www.coursera.org/learn/ai-for-medical-diagnosis

2) github.com/vishrutskaushik/Chest-X-Ray-Medical-Diagnosis

 

 

 

 

Chest X-Ray Medical Diagnosis with Deep Learning - ⑤ Prediction and Evaluation

end

BioinformaticsAndMe