PMLE Architecting Low-Code ML Solutions Practice Question

A hospital's radiology department wants to build a model that flags possible pneumonia on chest X-rays. They have 8,000 labeled DICOM studies in a Cloud Storage bucket and no in-house data science staff. They need a managed service that can ingest the images, train a classifier, and provide an endpoint for their viewing software. What should they do?

⚠ Common exam trap

The trap here is reaching for a general Vision API because it 'sees images,' when it cannot be trained on the department's pneumonia labels.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Create a Vertex AI dataset from the images and train an AutoML image classification model

AutoML image classification is the managed, low-code route for a labeled image set: it reads images from Cloud Storage, trains and tunes a classifier, and serves predictions through an endpoint the viewing software can call. The alternatives either use fixed-label APIs, skip the essential fine-tuning, or apply a streaming analytics product instead of a diagnostic classifier.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Use Vertex AI Vision to build a pipeline that counts objects in the X-ray streams
Why it's wrong here
Vertex AI Vision targets video and image stream analytics such as object detection and tracking, not diagnostic classification of static radiographs. Configuring it here would not produce a pneumonia classifier and misapplies a streaming analytics product to a diagnostic task.
✗
Use the Cloud Vision API label detection feature on each X-ray
Why it's wrong here
Vision API label detection returns general object and scene labels from a fixed model and cannot be fine-tuned on pneumonia-positive versus negative X-rays. It would produce generic labels like 'medical imaging' rather than a clinical pneumonia probability the department can act on.
✓
Create a Vertex AI dataset from the images and train an AutoML image classification model
Why this is correct
AutoML image classification accepts labeled images referenced from Cloud Storage, trains a managed classifier, and deploys an endpoint callable by the viewing software. It requires no model code, fitting a radiology team without data scientists who need a fast, managed path to a pneumonia flagger.
✗
Deploy a pre-trained TensorFlow Hub model to Vertex AI Endpoints without retraining
Why it's wrong here
A generic pre-trained image model was not trained on chest radiographs and would not reliably separate pneumonia from normal findings. Without fine-tuning on the 8,000 labeled studies, its predictions would be clinically meaningless, so this path fails the department's accuracy need.

Go deeper

PMLE Architecting Low-Code ML Solutions Practice Question

Why each option matters

Option-by-option breakdown

Related to this question