I was very excited about reviewing this book and writing what I thought would be a great summary on the advances of deep learning in medical decision support. However, the text is so poorly written that it is almost impossible to read or make sense of it. Either it is composed of disconnected shallow sentences or has serious grammatical shortcomings. The organization of the book is also strange and does not allow readers to build on and accumulate previous knowledge; instead, it feels like a collection of random thoughts to random citations (often misplaced).

The authors try to provide an overview of some of the works in the deep learning field, but there is neither depth nor a critical view or coverage of the limitations of these statistical approaches (deep neural networks). The result instead is like an attempt at a commented bibliography.

The authors often use the book’s length (very short) as an excuse for the lack of depth. It is really a very hard text to read or use in any useful way. I hope they can rewrite it. As it is today, as disconnected as it appears from sentence to sentence, paragraph to paragraph, and chapter to chapter, it feels like a book written by a suboptimal deep neural network.