A non radiologist must have the basic knowledge about drawing conclusions from an USG / CT/ MRI report.

Personally I have realised that there does occur certain amount of discrepancies in what the radiologist comments and what a clinician opines.

Thus it is imperative that we as clinicians must be able to draw inferences from a radiological imaging!

My opinion would somewhat differ. I routinely see patients referred to us with significant morbidity and mortality from clinicians erroneously drawing their own conclusions about imaging. This VERY frequently occurs where complex MRI artifacts that are easily recognized by a radiologist trained in MRI physics are misinterpreted by clinicians, along with other routine mistakes I see. I am fortunate that my clinical teams frequently are in communication about imaging findings and expressing their concerns all for the benefit of our patients. There once was a time before PACS where you would have had to go to the radiologist to even see the images! This forced open communication. I would STRONGLY suggest to you that if you have concern about a radiologist report, you should DISCUSS WITH THEM! Both the clinician and the radiologist bring an enormous amount of unique knowledge and experience to the table.

" experience is a bad teacher "

As for as conventional Radiology, CT and MR concerned, one can comment on images after radiology training but for US images its really difficult relaibly on images as its a dynamic study one have to relay on writen report by study performing expert.

Best suggestion for reffering clinicians is to discuss with Radiologist if there is some imbiguity or quary regarding report instead of making your own opinion. Some Neuro surgeons and chest surgeon are really found very good in imaging interpretation.

I take your brief question as general in nature, and so offer a general answer. Since my reply is general, there will be exceptions. In general, it is the obligation of the requesting provider to obtain and read the entirety of the report created by the diagnostic radiologist. Should elements not be understood, it is the obligation of the requesting provider to make himself or herself clear on those elements; sometimes, this is done by asking the/a radiologist. Finally and more directly, there is no little or no obligation for a non radiologist clinician also to know how to read the images themselves, nor is there any significant obligation actually to do so. In the U.S., diagnostic radiology training takes a minimum of four years (after internship) of intensive, comprehensive study, experience and examinations; it is similar in Canada, Europe and elsewhere. It would be unfair to expect non radiologists to know what radiologists know; in my experience, the finest of non radiologist clinicians are very well aware of that! Yes, as they become more experienced in their specialty and perhaps subspecialty, they become better at interpreting imaging exams relegated to that narrow focus. But let us never forget the classic example: an ED (or rheumatology or orthopedics, etc.) patient has shoulder radiographs for shoulder pain. Findings present or absent at the shoulder might indeed be properly noted by the ED (or rheumatology or orthopedics, etc.) clinician; but can we expect him or her also to note the lung nodule present at the corner of the image, since the lung is so close to the shoulder? Of course, that lung nodule might be far more important!

Rule 2: "It is easy when it is easy and difficult when it is difficult"

I would agree to the point that there should be basic knowledge of imaging even for a non-radiologist. But would it be apt to expect a "radiologist" to know about all other fields, in his or her short training curriculum?.... If "NO" then why would the converse be expected?

the "win-win" situation would in all probability be either working in unison, or to sub-stratify even the radiologists as "Onco-radiologists" or "URO-radiologists" etc.