We use this in the clinical setting when our standard PCR tests, cultures and serologies come back negative. This test is better for when we have exhausted more specific tests. Having such a broad sequencing of dna in the blood is not clinically useful because you may not be sick but be positive for quite a lot of things. Blood isn’t sterile. What would we do about it? Unclear, probably nothing, unless we have associated clinical findings/symptoms that correlate. Also a very expensive test.
Another medical student here, also with a very similar ML/programming background. Like the other commenter, I couldn't agree more with your analysis of the main issues plaguing current DL applications in medicine. When I was working on deep learning projects as a undergrad before coming to medical school, I naively assumed that solving a simple image classification problem for cancer detection would be enough. However, as one learns in medical school, imaging is only one component of an entire clinical vignette. Even before a patient undergoes a specific test, the history and physical exam really drive the initial protocols. Sometimes, without having this background knowledge, the classification from a simple program has no utility. A radiologist or pathologist typically has access to this information and can interpret this in the context of the clinical suspicion being put forth. I still believe AI has a role to play in easing this workflow, but "replacing" physicians will take a lot more than detecting some disease "better" or more "accurately" than a physician.
I would hesitate to write that "too mucus" is not an issue, because too mucus can cause a lot of problems for your lungs. For example, individuals with chronic bronchitis have hypertrophy of the mucus glands relative to the bronchial wall (look up Reid's index). This leads to mucus plugging within smaller alveoli and subsequent cyanosis and hypoxemia. Moreover, the obstruction due to the stasis of mucus can lead to the formation of purulent material/inflammatory necrosis (bronchiectasis) as well as damage to the mucociliary escalator. In fact, this mucus trapping is an underlying cause of the respiratory symptoms you see in cystic fibrosis and several other respiratory diseases. *I would add that in CF, the formation of the mucus is fundamentally impaired, causing it to be less effectively cleared.
Adding to some of the other comments: Leukocytes (immune cells) such as B and T cells will undergo "education" when they are developing. For example, B cells, before being released, will go through something called central tolerance in the bone marrow. If the B cell receptors bind a "self" antigen (markers that are associated with our own cells) while developing, it will go through negative selection and be destroyed. Of course, some B cells that bind self antigens do make it through in rare cases and this is what manifests as an autoimmune disease.
We use this in the clinical setting when our standard PCR tests, cultures and serologies come back negative. This test is better for when we have exhausted more specific tests. Having such a broad sequencing of dna in the blood is not clinically useful because you may not be sick but be positive for quite a lot of things. Blood isn’t sterile. What would we do about it? Unclear, probably nothing, unless we have associated clinical findings/symptoms that correlate. Also a very expensive test.