Patients with heart failure and cardiomyopathy that have been referred for Cardiac Resynchronization Therapy (CRT) will have their CRT devices implanted using an image guided treatment plan for optimal lead delivery. The validation of image-guided CRT lead delivery will significantly advance the field of heart failure therapy.

Objective 1: Using established study data from our previous study (Multimodality imaging Assessment of Pacing InTervention in Heart Failure Heart Failure or MAPIT HF) define characteristics of the "optimal myocardial target" that maximally yields clinical response to CRT.

Objective 2: Using established MAPIT-HF study data identify whether alternate coronary venous pathways to "optimal myocardial targets" exists in clinical non-responders.

Objective 3: Evaluate the feasibility of guiding the delivery of CRT leads to "optimal myocardial targets" through the use of an interactive, 3D "integrated cardiac model" of coronary vein anatomy, myocardial scar and mechanical dyssynchrony.

Currently there is no standard definition for response to CRT. Of importance is that measures of clinical response appear to be incongruent with echocardiographic parameters of reverse remodeling following CRT. The primary endpoint of clinical response will be an improvement in Left ventricular end systolic volume (LVESV) by ≥ 15% at 3 months following CRT, as validated in prior studies. Secondary clinical endpoints will also be evaluated; 1) Improvement in ejection fraction ≥ 5 percentage points, 2) Improvement in 6-minute hall walk ≥ 30 meters or ≥ 10%, 3) Improvement in NYHA (New York Heart Association) functional class by 1, or Specific Activity Score by 1, and 4) Improvement in Quality of Life (Minnesota Living with Heart Failure) by 10 points.

Our hypothesis is that a single 3D dataset can be developed from multiple MRI datasets for the simultaneous display of coronary venous anatomy, myocardial scar, and mechanical dyssynchrony, the latter being derived from tagged MRI data. The investigators hypothesize that the availability of this 3D model during fluoroscopic CRT lead placement will result in improved rates of lead tip delivery to "optimal myocardial targets" and improved clinical response. Our specific hypotheses of Project 2 are as follows;

1. The generation of an integrated cardiac model of venous anatomy, myocardial scar and dyssynchrony is feasible using multiple MRI-derived datasets.
2. The availability of this interactive 3D cardiac model during fluoroscopic CRT lead delivery will increase the occurrence of CRT lead tips being delivered to "optimal myocardial segments" when compared to historic control data within the MAPIT-HF study cohort.

Inclusion Criteria:

• any patient (> or equal to 40 years old) referred for cardiac resynchronization therapy
• New York Heart Association (NYHA) greater than or equal to 2
• ejection fraction </=35%
• QRS duration >/=120 msec
• Patient has been on stable heart failure medications for at least 6 weeks and in the investigator's opinion reached optimal medical therapy for treatment of heart failure over the past 6 months

Exclusion Criteria:

• patients with contraindications to MRI
• recent myocardial infarction within last 6 weeks
• cardiac revascularization procedure within the past 3 months
• glomerular filtration rate (GFR) \