Written by Laurence Moroney, AI Technology Advocate
We know that machine learning technology can be used to solve various scenarios (such as healthcare) important tasks, and educators and experts in various fields are guiding and training developers on how to use machine learning technology to solve real-world problems. We are excited to share with everyone an important development from the TensorFlow community: deeplearning.ai and Coursera have jointly launched the “AI for Medicine” specialization course based on TensorFlow.
What we look forward to the most is seeing people use TensorFlow to solve some of the problems they encounter in real life.
The AI for Medicine specialization course includes three courses, taught by Andrew Ng, with Pranav Rajpurkar as the main instructor. We hope more people will participate in this course to understand the demand for machine learning in the medical field.
Deeplearning.ai and Coursera designed this specialization course, which includes the following three courses:
The first course: Using Machine Learning for Medical Diagnosis, introduces you to some hypothetical machine learning scenarios to understand how to diagnose medical issues. In the first week of the course, you will explore scenarios such as detecting skin cancer, eye diseases, and tissue pathology. You will practice how to write code using convolutional neural networks in TensorFlow to view images, which have various applications, such as identifying different health conditions using X-rays.
This course requires you to have a certain understanding of TensorFlow and will use convolutional neural networks, transfer learning, natural language processing and other techniques. It is recommended that you first take the TensorFlow: In Practice specialization course to learn the coding skills behind it, and participate in the Deep Learning specialization course to gain a deeper understanding of how the underlying technology works. Another important resource for understanding the technologies used in this course is Aurelien Geron’s book Hands-On Machine Learning with SciKit-Learn, Keras, and TensorFlow.
One of the highlights of the course is that it does not overuse medical terminology; we use general machine learning techniques in TensorFlow (such as data augmentation) to improve the model (Note: All data used in this course has been anonymized).
Rajpurkar and Ng’s course exercises: Using image augmentation to expand the effective size of the dataset
This course will continue to use evaluation metrics and isolate key points, and introduce how to accurately interpret confidence intervals.
At the end of the first lesson, we will once again delve into image processing, this time using segmentation on MRI images to complete the programming task of automatic segmentation of brain tumors on MRI.
The content of the second lesson in the specialization course is Machine Learning in Medical Prediction, where you will learn to predict patients’ future health status by building models. You will also learn techniques for extracting data from various reports such as patient health indicators, health history, and demographic information to predict the risk of patients developing major diseases (such as heart attacks).
The third and final lesson is Machine Learning in Medical Treatment, which will use models to assist healthcare teams in predicting the impact of medical treatments on patients. It will also involve using machine learning to process text and learn to extract information from X-ray radiology reports using NLP techniques, obtaining labels or foundational knowledge for robots to answer medical questions.
In Andrew Ng’s words, “Even if you are not currently working in the medical field, I think you will find these application scenarios and their practice very practical, and perhaps this specialization course will spark more interest in the medical field.”
This specialization course is available on Coursera and, like all courses, can be participated in for free. You can learn more about deeplearning.ai on the deeplearning.ai website, and find more information about TensorFlow at tensorflow.google.cn.
If you want to learn more about the related content mentioned in this article, please refer to the following documents. These documents explore many of the topics mentioned in this article in depth:
-
deeplearning.aihttp://deeplearning.ai
-
Courserahttp://coursera.org
-
Course Resources
-
AI Medical Specialization Course (Read Original)https://www.coursera.org/learn/ai-for-medical-diagnosis
-
TensorFlow: In Practicehttps://www.coursera.org/specializations/tensorflow-in-practice
-
Deep Learninghttps://www.coursera.org/specializations/deep-learning
-
Hands-On Machine Learning with SciKit-Learn, Keras, and TensorFlowhttps://www.oreilly.com/library/view/hands-on-machine-learning/9781492032632/
-
Segmentationhttps://tensorflow.google.cn/tutorials/images/segmentation
-
Participationhttps://learner.coursera.help/hc/en-us/articles/209818613-Enrollment-options
-
TensorFlow Official Websitetensorflow.google.cn
For more information, please click “Read Original” to visit the AI Medical Specialization Course.
