As genomic data has enabled more precise cancer treatment, the emergence of Artificial Intelligence (AI) has further revolutionized oncology.
By integrating computer science with medical practice, AI allows for more accurate diagnosis, more effective treatment, and reduced workload for healthcare professionals and systems.
Before exploring how AI transforms cancer care, let’s first understand what artificial intelligence really is.
By Bryant Chen, Medical Affairs Manager, Be Accelerator
What Is Artificial Intelligence (AI)?
In today’s world, AI surrounds us—in e-commerce platforms, social media, and even transportation systems.
At its core, Artificial Intelligence refers to the capability of machines—through the integration of software and hardware—to interact with humans and interpret the world.
AI can generally be categorized into two main types:
-
Strong AI:
Refers to machines with self-awareness and independent thought, capable of human-like reasoning and creativity.
Examples often appear in science fiction, such as Jarvis from Iron Man, an AI assistant that anticipates needs and executes complex tasks—sometimes even beyond human intuition.
However, true strong AI remains theoretical and does not yet exist in reality. -
Weak AI:
The form we encounter today, capable of responding intelligently to specific tasks using data-driven logic and algorithms, yet lacking self-awareness.
Weak AI systems can outperform human decision-making in certain scenarios—but they do not “think” independently.
Currently, the realization of AI primarily relies on a method known as Machine Learning (ML).
Machine Learning (ML)
Machine Learning enables computers to “learn” from large amounts of data through mathematical algorithms, allowing them to improve accuracy over time by detecting patterns and reducing errors.
As data diversity and computational demand have grown, multiple ML techniques have been developed and widely adopted across domains:
• Classification
Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), Naive Bayes Classifier, Gradient Boosting Tree, Neural Network.
• Prediction
Linear Regression, Decision Tree, Random Forest, Gradient Boosting Tree, Neural Network.
• Dimensionality Reduction
Principal Components Analysis (PCA), t-Distributed Stochastic Neighbor Embedding (t-SNE), Uniform Manifold Approximation and Projection (UMAP).
• Clustering
K-Means, Hierarchical Clustering, Gaussian Mixture Model (GMM).
• Association
Apriori Algorithm for rule mining and correlation detection.
In most ML applications, algorithms rely on feature extraction—structured information used for training.
However, in complex domains such as computer vision, traditional ML often struggles to reach human-level accuracy.
This limitation has led to the evolution of a powerful subfield: Deep Learning (DL).
Deep Learning (DL)
Deep Learning is a specialized branch of machine learning, inspired by the biological neural networks of the human brain.
Data is passed through multiple “layers” of artificial neurons, each with assigned weights that represent how strongly a feature contributes to the final decision.
By processing errors layer by layer, the network refines its understanding—eventually producing highly accurate results.
Today, deep learning powers applications such as computer vision, image recognition, speech processing, and object detection—enabling medical AI systems to detect tumors, interpret radiological scans, and predict disease outcomes with remarkable precision.
Where Is Machine Learning Used?
Machine learning is now widely applied across industries:
automotive, manufacturing, entertainment, retail, e-commerce, and finance.
In healthcare and life sciences, the abundance of medical and health data has made ML indispensable for disease prediction, early diagnosis, and treatment optimization.
In our next article, we’ll explore how AI specifically integrates with cancer medicine, transforming patient outcomes and redefining clinical precision.
For more information, please contact:Bryant.chen@be.tworg.app
