AI-900 Question 150
Select 3A healthcare organization wants to use computer vision to automate the identification of fractures in X-ray images. Which types of computer vision solutions are most appropriate for this use case?
- A
Image classification
- B
Object detection
- C
Optical character recognition (OCR)
- D
Semantic segmentation
- E
Face recognition
Show answer and explanation
Correct answers: A, B, D
Explanation
For analyzing X-ray images to detect fractures, image classification, object detection, and semantic segmentation are the most relevant computer vision solutions. Image classification can determine if a fracture exists, object detection can locate specific areas of interest, and semantic segmentation can outline the fracture with greater detail. OCR and face recognition are unrelated to this medical imaging scenario.
- A. Correct.
Image classification is suitable because it can assign a label to an entire image, such as 'fracture detected' or 'no fracture detected'.
- B. Correct.
Object detection is appropriate because it can identify specific regions within an X-ray image where a fracture may be present.
- C. Incorrect.
OCR is not relevant here because it is used to extract text from images, which is unrelated to analyzing medical scans.
- D. Correct.
Semantic segmentation is suitable as it can provide pixel-level classification, helping to highlight the exact areas of the fracture in the X-ray.
- E. Incorrect.
Face recognition is not relevant because it is designed to identify and verify human faces, which does not apply to X-ray analysis.