Submission 2025
| Submitted by: | Sharanya Balachandran |
| Department: | Radiology + Diagnostic Imaging |
| Faculty: | Medicine + Dentistry |
The image depicts the need of echo cardiography fusion. When cardiac ultrasound is obtained from humans from a single point of view, some of the anatomical structures are hidden and/or not visualized clearly. In order to obtain the complimentary details, the images acquired from multiple views can be fused. The aim of the research is to improve the diagnostic quality of cardiac ultrasound images using fusion techniques.
I am developing a machine learning based fusion network to fuse the cardiac ultrasound images obtained from multiple views of the heart. I aim to improve the diagnostic accuracy of cardiac ultrasound images to various heart diseases by enhancing the quality with the datasets acquired from robotic arm and collaborating with the cardiologists.
Was your image created using Generative AI?
Yes.
Did you upload any materials into an AI tool to help generate your image?
No.
Which model of generative AI did you use?
DeepAI.
How was your image created?
The image was created using DeepAI, using prompts as input, and it was enhanced using the options provided by DeepAI. This tool converts written descriptions given as prompts into intricate visual representations by interpreting them. The image was generated by giving the suitable prompts and then enhanced with the built-in tools to improve the aesthetic appeal.
How did you craft your prompt to guide the generative AI in creating your image? Please describe the technique or approach you used to convey your vision to the AI.
The aim was to generate an image that illustrates the need for echocardiography fusion using DeepAI. I provided the starting prompts with setting a background for the image by giving a prompt, for example: “Generate an anatomical view of a 3D heart with cardiac ultrasound image as the background,” aiming to visualize the image related to echocardiography. Then I gave detailed prompts to generate this image by refining my input texts. To exhibit the fact that some of the cardiac structures are not visualized from a single point of view, I added images of humans looking at the heart from different angles, which also added illustration to the need for fusion of images.
Where is the image located?
This image is generated using DeepAI with the prompts.