The videos were too short and seem to be missing some critical information that was included on the tests. The assignments, from a program structure perspective, were hard to follow. Lastly, I was a little disappointed on the lack of AI concepts included.

This course was great and more challenging that I have expected. More focus on statistics and survival data which is important for prognosis. Course has a good flow and valuable content.

The course builds upon the statistical formulations for various prognostic models . As far as I experienced, the course requires no background in deep learning or machine learning and neither in medicine, so if you're interested in the topic you don't have to worry about any pre-requisite. Unlike other courses you will not be using neural networks or deep learning algorithms, however the assignments uses a lot of real life data for prognosis tasks, so it's still an exciting course to build upon you own prognostic models.

Excellent course! Real world data and robust models. Of particular value was the implementation of the SHAP feature interpretation algorithm as applied to ensemble models.

Excellent course that teaches you complex concepts in an easy to understand manner. The assignments are challenging, while at the same time teaching you advanced stuff!

Thank for this course, I really enjoyed it! It was well structured and the explanations are very clear! The labs are amazing, as they really help understanding the models and how to use them. Some recommendations:- I really like the way that the course is structured and that you are beginning with naive and easy solutions to come up with better ones, but sometimes it wasn't explicit enough that they where naive solutions while you were presenting them. Therefore it was a bit frustrating because during the first videos, i was always like "Oh man why is he doing that? I have a better way!"... only to realize that you were about to present something better afterward - Too many pure computation questions in the quiz... I find it useless to have questions with a formula and the data and we only need to understand how the formula is applied... It is too easy and it doesn't help understand deep meanings and explanations.- The quizzes are not mandatory to pass the course and they should be!- When a model is presented in the video, you could also explain how it is going to be trained, because sometimes i didn't know if some weights that you were talking about were learned or were just hyper parameters (ex: in basic models exemples of week 1)Another example is at final week, in the video 'survival tree', you manually split the data but you don't explain how the survival tree will be trained, now that we don't have a labelled outcome 1 or 0 associated with each patient.- It is not stated if the models and solutions that you are presenting are the state of the art, or if we should investigate more.- It would be nice, for each course, to have an exemple of solution that is already used by some hospitalsDon't hesitate to reach out to me if some of my recommendations are not clear enough... My english isn't perfect :) Thanks again, i'm looking forward to do the last course!

Really helpful course if one wants to understand the application of machine learning in the field of medical prognosis,i.e., to predict the medical condition from the given data. The instructor is really good at explaining things and the videos as short, crisp, and covers all major aspects needed to be understood by the student. I would recommend to undertake this course if you want to work in the medical applications of ML and familiar with the basics of ML.

A well structured courses, which is very fun to go through. The lectures are very short and the learning is more practical and application oriented. The tutor is very knowledgeable and navigates us through some of the tough concepts in a logical way. The course is free from medical jargons and thus is very recommended for CS people interested in applying their knowledge to medicine, without any prior knowledge of Medicinal jargons

This course is one of the best courses to learn about Medical Prognosis. Really, the survival models were described in great detail. Thank you, Pranav for this wonderful course.

This was a wonderful course, I am now able to see how this course relates to the actual medical field, where we go from diagnosis to treatment and finally prognosis.

It seems to me the course is at least half about statistics, not AI. Nevertheless, the content is interesting and well explained. The coding exercises and projects may makegood starter code for your own projects.

It is clearly explained, there is even too much hand-holding. When the instructor shows a (very basic) integral and reassures us that even if we don't know what it is, we will be alright, he is pushing it too much. Anyone who hasn't seen an integral like that before is not in the audience of the course anyway.

The projects are very interesting and relevant, it is a pity they have been dumbed down in the execution to the point that we are required to fill in missing lines at specific places. I believe ungraded exercises should build up to the project, but then for the project we should have more free reign. Again, beginning programmers have no business attending this course anyway.

Awesome course! Anyone looking to get an insight into how to perform studies which are preemptive, especially in healthcare must go through this course. The instructor is lucid, delivers content fantastically and there's sufficient supplementary resources as well. Only thing which I would have liked more is more practise assignments. Otherwise for content as a whole, it's a 5/5.

the coding assignments were not that hard sadly, but the knowledge about techniques and methods and formulas to interpret the prognosis is very helpful

Some more support about the assignments in the forum or in the slack channels could be really helpful to better understand the exercise solutions.

This is a nice introductory course for some of the machine learning applications in medical prognosis. However, it lacks depth in ML algorithms and it focuses more on data engineering and preparation for medical applications. As a tip try to read some of the citations or google the concepts (random forests, ...) in medium or other data science blogs.

Great course. However, i miss how Andrew deconstructs everything - it completely absorbs all your curiosity. When you move to the assignments, without extra work you can fully understand how the libraries work. This however has a different approach, they absolutely open your mind up and enthuse you to do much more background work. really good stuff!

I feel that video and other materials are not preparing well for the quizzes.

Thank you Pranav Rajpurkar and Andrew Ng for this amazing specialization! Thank you deeplearning.ai! Thank you Coursera!This specialization covers application of AI algorithms for: medical diagnosis of patients using chest X-Rays and 3D MRI brain images; prognosis of patients using survival models; and medical treatment recommendation models.The lectures were brief and comprehensive, the quizzes included toy problems to test the grasp over the mathematical formulas, and the assignments were simple and covered implementation of most of the concepts taught in the courses.

We'll its not definitely an Andrew Ng level of simplifying complex topics and explaining the intuitions behind. But the author Pranav did a great job on compiling useful topics that will sort of guide the practical and cutting edge (i.e. SHAP library) applications of AI in Healthcare. I used it as a curriculum guide on which topic to further study and deep dive. Each videos was just like 1 minute high level explainer for a topic, but it should give you a head start on what to focus on. Overall, I this is very satisfactory!

This course is as good as the AI for Medical Diagnosis. It is clear all along, very well illustrated, the notebooks are extremely well done. The mathematical concepts are quite easy and well explained, making this course accessible for almost anyone. The AI part is a bit more limited than the other course, but the estimators and evaluation methods (Kaplan-Meier, Harrell's C index) are definetely important concepts as well to know to work in the medical area.

Thank you for this awesome course!

Absolutely fantastic course! Believe me you'll never get disappointed of a course/specialization by the "deeplearning.ai" team, the same goes for this one too. Pranav has done an excellent job explaining the intricacies involved with medical data and has very well extrapolated these in accordance to the deep learning concepts.

This course teaches an individual about the different prognostic tasks in medicine along with the machine learning models that can be used. I really enjoyed this course as it taught me a lot about how one can create survival and risk models using linear models and decision trees. Another set of skills added to my toolkit!

The medical prognosis is a comprehensive course for predicting and analysing future condition of patient for medicine. Highly recommended for anyone with interest in medical research. Even if you don't understand the concepts, re-visit it again. This is called learning with patience as 'patient'.

This course is well organized and have a good flow, that helps to understand all the facts. The assignments are also good and have a lot to learn. Thank you very much for providing a platform for students to learn this area, AI in medical applications.

Great course! I really enjoyed the part of the illustration of how to handle the right-censored patients and have understood how the chain rule of probability is applied. It would be better if the python assignments have more exercises for practice ;)