 Once you've identified the lesion, we have to be sure, is it a mediastal lesion or is it something else? The two main lesions we need to differentiate form is pulmonary lesions and cardiac lesions. So mediastal versus pulmonary, there are few points which may help us to differentiate. The mediastal lesions have a smooth contour, they usually form an obtuse angle with the surface. There is absence of the air bronchogram, it may be seen extending bilaterally. All the mediastal lines that we described as till now, they may disrupt those and they may have associated spinal, costal and sternal abnormalities. Whereas lung lesions are usually surrounded by air, they form an acute angle with the surface, it may contain air bronchograms and usually unilateral. So this is how a mediastal mass and a lung mass will look schematically. The mediastal mass sits under the surface of the mediastal and it creates an obtuse angle with the lung. This is a lung mass which abuts the mediastal surface and creates an acute angle with the lung surface. Now again, as you can see there is a mediastal mass which is allowing the right mediastal border and this turned out to be a thymoma. This is a lesion which is seen in the right upper zone which is forming an acute border with the mediastal number. This is a lung mass which turned out to be a pancostum. This is how we differentiated mediastal and pulmonary lesions. Now to differentiate between mediastal and cardiac lesions, we have the hyalum ovuli sign and the hyalum convergent sign. Now the hyalum ovuli signs says that if you can visualize the hyalur vessels through a mediastal mass, then the mass does not arise from the hyalum. So as you can see this is a large mass out here and we can see the pulmonary hyalur vessels through the mass. So this does not arise in the region of the hyalum. Because of the way the mediastal is made, most of these masses will be located in the anterior mediastal. This same lesion which is seen on a CT scan which turned out to be a lymphoma and an HIV positive patient. Again, the hyalum convergent sign, this is basically to distinguish a bulky hyalum due to pulmonary artery hyperplasia from a juxtahyla mass or a lymph nodal alarm. If the pulmonary artery branches are visible through the opacity and these branches converge towards the waist of the heart, that means they are going inwards through the heart. This is a positive hyalum convergent sign and this opacity is likely to be due to a juxtahyla mediastal mass. However, if the pulmonary artery branches lead towards the abnormal bump, then the opacity is likely to be due to an enlarged pulmonary artery which is called as the negative hyalum convergent sign. So once we've decided that there is a lesion and it is a mediastal lesion, it is important to put it into a compartment. There have been many classifications and divisions. The anatom is divided into four parts. There is a classification which is mainly on the lateral chest area given by Felsen which we follow which divided into three parts. But it is important to note that these divisions are all theoretical rather than physical and that the disease can spread from one compartment to the another and that some diseases do not exclusively occur in any one compartment. Let's look at the Felsen's classification. These are basically based on findings of the lateral chest radiograph. There are two lines we must draw. One line is extending from the diaphragm to the thoracic inlet along the back of the heart and anterior to the trachea. It separates the anterior and the middle mediastal. So everything which is anterior to the line is the anterior mediastal. Then the second line is a line that connects points one centimeter beyond the anterior margin of the vertebra. This separates the middle and the posterior mediastal. So between the first line and the second line is the middle mediastal. Behind the second line is the posterior mediastal. On the radiograph to differentiate into which compartments we have a few signs. The silhouettes sign, the cervical thoracic sign and the thoracobdominal sign. Now the silhouettes sign says if an intrathoracic region is in anatomic contact with the heart or the aorta or the diaphragm, it will obscure that border of the silhouette if it has the same density. This is well demonstrated here. There is loss of the cardiac silhouette with the border of the heart on the right side with relatively low density. The CD scan showed an epicardial, prominent epicardial fat path, which is seen on the right side. Once we've decided there is a lesion, there is a mediastal lesion, which compartment it is, then it is also important to narrow down the referential diagnosis depending upon the content of the lesion. So basically we should look at if there is fat, is there a calcification or is there fluid? Now these are the fat containing, common fat containing lesions, thymolipoma, teratomas, usevigilipomas and so on. This is an anterior mediastal lesion which contains macroscopic fat. There are some cystic areas and there are some solid areas also. This turned out to be a teratoma. These are the lesions which contain calcific focac, dense areas of calcification. As you can see on the frontal and the lateral radiograph, you can see a rim calcified lesion. Rim calcified lesion within the anterior mediastal. On the CD scan again this turned out to be an enhancing rim calcified lesion in the anterior mediastal which was diagnosed as a thymoma. Now what are the fluid containing lesions? The common ones in the anterior mediastal thymic cyst, then we have the thymomas, teratomas, pericardial cyst, focac duplication cysts and so on. This is a simple unilocular cystic lesion which is seen in the anterior mediastal, this turned out to be a thymic cyst. Now more important also is after the contents, does the lesion enhance or does it not enhance? The enhancing lesions are obviously hyper-enhancing lymph nodes, thyroid tissues, paraganglionomas, hemangiomas and any vascular lesions. As you can see here, a descending thoracic aortic aneurysm. In addition to this we need to see whether there are any associated abnormalities. If there are presence of localized pulmonary lesions in addition to the mediastal abnormalities as occurs in malignancies and tuberculosis, if there is any abnormal, abnormal involvement in additional thoracic, as happens if there is a rare sewer system in the thorax, neuroblastomas, and if there are any associated vertebral anomalies such as neuroendrixes and neurogenic tumors. So most of this diagnosis can be made up to the content of the lesions on chest radiography. But what is the role of computer tomography in a mediastal lesions? To define which compartment is it formed? Yes, to confirm which compartment. What is the content of the lesion? What is the enhancement of the lesion? Any associated abnormalities? And based on the division and description, ultimately we need to give a diagnosis based on the CD findings. There were different classifications by anatomists, there were different classifications by surgeons and different classifications by radiologists. So there was a need to have a classification based solely on cross-sectional imaging because that is what gives us the final diagnosis. So this international thymic malignancy interest group came out with a classification in roughly around 2015-2016 based solely on the cross-sectional findings on CT. They divided the mediastal into three compartments. One is the pre-vascular compartment, which is basically bounded superiorly by the thoracic inlet, inferiorly by the diaphragm, anteriorly by the sternum, and posteriorly, which is important to note, by the anterior aspect of the pericardium as it traps around the heart. Okay, the contents include thymus, fat, lymph nodes, and the left pre-chrysophallic vein. The visceral compartment was anteriorly bounded by the posterior boundary of the pre-vascular compartment and posteriorly by the vertical line connecting a point on the thoracic vertebral bodies, one centimeter posterior to the anterior margin. And the paravortable compartment was by the posterior body of the visceral compartment and the posteriorly by a vertical line against the posterior margin of the chest wall. So this is schematically how this division looks like. Pre-vascular compartment outlined in red. The visceral compartment outlined in blue and the paravortable compartment outlined in yellow. These are various sections at the level of the aortic arch, the left pulmonary artery and at the left atrium, which shows how the division of the mediastome runs through and through. Again, based on this division, we have certain divisions which occur in certain compartments and we can divide these based on the content of these lesions into solid cystic and fatty lesions. So as you can see in the pre-vascular compartment, there's a large lobular heterogeneous pre-vascular lesion with a large amount of macroscopic fat. There are some soft tissue within it and there are some fluid density with this. This turned out to be a mature teratoma. Again, there is a well-defined low attenuation lesion in the thymic bed in the pre-vascular region. This is unilocular with no evidence of any internal septa or soft tissue. This turned out to be a benign unilocular thymic cyst. In the visceral compartment, basically, you can divide it into solid and cystic lesions. Again, you can see a well-suckumscribed homogeneous subcaranel lesion, which is seen indenting the left atrium. This is a bronchogenic cyst. The paravirtual compartments, basically, most of the lesions which occur here, that is, 75% of the lesions are of neurogenic origin. These include peripheral nerve-shoot tumors such as schwannoma, neurofibromas, plexiform neurofibromas, and malignant peripheral nerve-shoot tumors, sympathetic ganglia tumors, and paraganglia-paragangliroma tumors. The non-neurogenic tumors account for only about 25% of the lesions, which we can see. Again, these can be divided into solid cystic lesions. You can see that there is a well-defined heterogeneous soft tissue lesion in the right paravirtual mediastome, and on biopsy, this turned out to be a benign peripheral nerve-shoot tumor. Now, you must understand that you must see these lesions not only in the mediastal window, but also the bone window, because it is important to differentiate between benign and malignant lesions. The benign lesions usually cause just erosion, pressure erosion of the adjacent vertebra, but if there is destruction or invasion of the bones, then that goes more in favor of a malignant lesion. And MRI at this point in time also helps us to delineate whether there is any intra-spinal extension. So what is the role of MRI in mediastal lesions? One, it is the most useful in major modality if there is a confusion whether the lesion is cystic or solid. Example, if there are time-exists, to differentiate time-exists from solid time-exisms, whether there are any cystic or necrotic components to delineate septic within lesions, or soft tissue components within cystic lesions. Then for evaluating neurogenic tumors and extent into the spinal canal, MRI is also useful in those cases. In intra-cardiac tumors, it is helpful in delineating intra-cardiac tumors. In patients with renal failure in which we cannot give IV contrast, a plain MRI helps us also to delineate the vascular abnormalities. So in conclusion, CT is the modality of choice in evaluation of mediastal lesions. Issue content of the lesion is important in arriving at a diagnosis. Some lesions can be diagnosed based only on the imaging criteria, such as restrosol, goiter, fat-containing lesions like lipomars, thymolipomars, and purely cystic lesions. Most other lesions need to be correlated with clinical details, tumor markers, and histopathology. And the itmic division of the itmic classification helps us in identifying and precisely identifying mediastal abnormalities at cross-sectional imaging.