The purpose of this study is to utilize cardiac imaging data acquired as part of the standard of care for these patients, such as MRI, 3D echo, and CT, and existing 3D reconstruction protocols to assess the feasibility of using surgical planning in the treatment of patients with complex cardiac defects. The specific aims of the project are as follows:

1. Develop a protocol to reconstruct heart models from patient imaging data and perform "virtual" surgery on reconstructed 3D anatomy using appropriate, pre-existing patient datasets.
2. Use the developed protocol to prospectively plan and evaluate the possible surgical options for new patients.
3. Validate that the optimal virtual anatomy agrees with what was surgically implemented using post-operative patient scans, when available.

In the United States, approximately 1 in 200 babies are born each year with harmful congenital heart defects (CHD) that require some form of medical management. Often, these defects consist of holes in the septum (the walls between the heart chambers) and/or abnormal development of the heart chambers or major blood vessels. Surgery is the primary treatment course for many of these patients and, through the use of patches and artificial vessels, it is often possible to repair the defects and recreate the normal blood flow path through the heart.

These techniques are not always simple, however, and the surgeon must take great care not to harm the pumping function of the heart. In more complex cases, the surgeon must decide between multiple repair strategies that will have a major effect on the long-term health of the patient. It would be helpful in such cases for the surgeon to be able to assess the repair options prior to the operation using virtual 3-dimensional representations of that patient's anatomy. Having this ability would remove some of the uncertainty from the decision-making process by providing accurate predictions of post-surgical anatomy.

In fact, the technology exists to include such a surgical planning tool into the standard treatment course for these patients. Using 3D anatomical images, acquired from basic, techniques such as magnetic resonance (MR), computed tomography (CT), and echocardiography, engineers at Georgia Tech have the ability to build accurate 3D models of patient anatomy, such as the heart. Using these models with a state-of-the-art graphics manipulation tool, surgeons would have the ability to virtually operate on the patient and select the optimal treatment approach, as previously discussed. Similar techniques have already been developed and used to plan surgeries for a limited subset of CHD patients with a single ventricle physiology.

The purpose of this study is to further develop these techniques and apply them to a broader range of CHD patients. To do this, patients undergoing an appropriate surgical repair will be recruited to participate in the study. Images obtained from pre-operative scans will be used to build the anatomical model, which the surgeon will manipulate to test the different available options. By successfully testing and eventually implementing these techniques in the standard of care for CHD patients, the optimal approach for reconstruction will be implemented more frequently, and thus patient outcomes will improve.

Inclusion Criteria:

• any pre- or post-operative patient with a complex heart lesion treated at Children's Healthcare of Atlanta by Dr. Kanter or his team that is appropriate for surgical planning

Exclusion Criteria:

• upon review by engineers at Georgia Tech, the image quality of the acquired scans is deemed insufficient to reconstruct an accurate 3D model

Patients with complex congenital heart defects, such as Double Outlet Right Ventricle, requiring surgical intervention, who are being treated at Children's Healthcare of Atlanta.