| 1 |
What is the primary goal of drug discovery?
|
A) To increase pharmaceutical profits |
|
Drug discovery is there to make better decisions faster. |
In pharmaceutical industry, models are usually implemented in a result-oriented fashion to save resources and time. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 2 |
Which of the following is a common use of machine learning in drug discovery?
|
B) Predicting the biological activity of compounds |
|
ML is used to make better decisions faster and to accelerate the DMTA cycle of novel molecular entities. |
ML generates knowledge to improve and expand methods such as predicting the biological activity of compounds. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 3 |
What is a compound library in the context of drug discovery?
|
B) A collection of chemical compounds tested for biological activity |
|
A compound library in the context of drug discovery is a collection of chemical compounds tested for biological activity. |
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 4 |
Which of the following best describes QSAR models?
|
A) Models that predict the quality of scientific research |
|
QSAR models are a necessary component of numerous drug discovery projects. |
QSAR is quantitative structure-activity and property relationship and it predicts the quality of the scientific research. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 5 |
Why is data curation important in machine learning for drug discovery?
|
B) It ensures the data is accurate and relevant for model training |
|
ML model performance heavily relies on the quality of the experimental data used for training. |
They need accurate data in order to use it in drug discovery. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 6 |
What challenge does the heterogeneity of public datasets pose for machine learning in drug discovery?
|
E) It ensures models are always accurate |
|
To increase data set size, public data are generally pooled from multiple sources, which in turn increases heterogeneity. |
Merging data sources implies major efforts and bears the risk of biases, redundancies, and error accumulation. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 7 |
In machine learning models for drug discovery, why is model validation critical?
|
C) It tests the model's predictive accuracy on unseen data |
|
Model validation mimics how a ML model will be used in practice. |
For example, to predict compounds that have not been synthesized or measured to gain trust in the models to better understand it. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 8 |
How does the "design-make-test-analyze" (DMTA) cycle benefit from machine learning?
|
C) By making the cycle unnecessary |
|
The complete DMTA cycle is rarely fully executed. |
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 9 |
What does the term "model deployment" refer to in the context of machine learning for drug discovery?
|
C) Making a trained model available for use in making predictions |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 10 |
Why is the democratization of models and data science practices considered a key aspect in pharmaceutical industries?
|
C) It enables scientists from different domains to contribute to and benefit from shared goals |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 11 |
What is a key component in the design of AI systems for medical diagnosis that ensures adaptability to various cases of melanoma?
|
A) Static databases of previous cases |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 12 |
What aspect of AI application in dermatology is considered essential for improving the accuracy of melanoma diagnosis?
|
A) The ability to process large datasets quickly |
|
Much of AI application in medicine relies heavily on big data analysis. |
The most critical precondition of emerging AI developments in healthcare is the data availability needed to develop and train algorithms. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 13 |
In the development of AI for medical diagnosis, why is explainability considered important?
|
D) It enables physicians to understand the AI's diagnostic reasoning and trust its recommendations. |
|
Explainable AI means that the user should understand the outcome produced by AI. |
Explainability is important because it allows physicians to understand the AI's diagnostic reasoning and trust its recommendations. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 14 |
What represents a significant challenge in the human-AI collaboration for medical diagnosis, according to the document?
|
C) Managing the complexity of human-AI interaction |
|
AI must be combined with human use in order to achieve diagnosis. |
Dermatologists do not believe that AI can accurately diagnose a patient all on its own, so they must use their judgment to aid in a patient's diagnosis because although there are benefits, there are also risks if AI were to diagnose a patient all on its own. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 15 |
Why is scientific proof of AI's validity in the medical field crucial?
|
C) It builds trust and confidence among medical professionals in AI's recommendations. |
|
Dermatologists are skeptical of the accuracy of AI diagnoses. |
If there were scientific proof on AI's validity, they may have more confidence and trust in AI's recommendations to be able to allow them to diagnose patients on their own. |
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 16 |
Which of the following best describes the role of AI in the diagnostic process?
|
C) To provide support and augmentation to human decision-making |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 17 |
What is a primary benefit of AI in medical diagnosis mentioned in the document?
|
B) Providing differential diagnoses with probabilities |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 18 |
Why is the integration of AI into medical diagnosis considered complex?
|
C) It involves combining AI's capabilities with human expertise. |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 19 |
Which statement reflects the dermatologists' view on AI-generated predictions?
|
D) They are less accurate than traditional methods. |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 20 |
What is crucial for dermatologists to effectively use AI in diagnosing melanoma?
|
A) Relying solely on AI for all diagnostic needs |
|
|
|
5 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|