Memory shortage error in MATLAB GUI code

I am currently designing code to create a GUI using the Field II ultrasound simulation tool. In my code, I am encountering an error of memory shortage similar to the attached file, preventing the GUI from running. I suspected it might be a RAM issue, so I upgraded from 32GB to 64GB, but the error persists. However, when I write the same code as a script without a GUI, I do not encounter any memory shortage issues. I would appreciate it if you could provide insight into what might be causing this error. I have also attached the specifications of my PC in the attached file.

• error code :

• PC specification :

1. CPU : AMD RAYZEN5 - 5th Generation 7500F
2. Mainboard : GIGABYTE B650M
3. Memory : DDR5-6000 CL30 TRIDENT Z5 RGB J (64GB(32GB*2))
4. Graphic card : RTX 4070 SUPER STORM X Dual D6X 12GB
5. SSD : SK hynix Gold P31 M.2 2280 NVMe SSD 1TB

[Steven Lord]

Please show the full and exact text of the error message (all the text displayed in red in the Command Window.) The exact text may be useful in determining what's going on and how to avoid the error.

[soomin hwang]

Thank you for your response. I have reattached the photo for clarity!

Answer 1

[Rupesh]

Hi soomin,

From the error code you mentioned I understand that you're encountering an "out of memory" error in your MATLAB GUI application, specifically when using the Field II ultrasound simulation tool. This issue arises during certain operations that might involve intensive computational tasks or large data handling, which are not present when running the equivalent code as a script. Based on the error code, you have given I can think of four possible issues contributing to this problem, along with a detailed workaround for each.

The workaround provided is based on common issues and best practices. However, if the script or specific segments of the code were available for review, a more pinpointed or exact solution could be possible.

1. Excessive Memory Usage in Data Processing

Issue: Operations like “calc_p” and “yz_beamfield” may involve processing large matrices or arrays, leading to high memory consumption.

Solution: Optimize these operations by reviewing the algorithms for efficiency. Consider using more memory-efficient data types (e.g., converting `double` precision data to `single` if high precision isn't crucial). MATLAB's `sparse` matrices can also be used if the data contains a lot of zeros. Regularly use the `clear` command to free up memory by removing variables that are no longer needed.

2. Memory Leaks from Persistent References

Issue: Memory leaks can occur if the GUI holds onto references to large objects or data arrays that are no longer needed, preventing MATLAB's garbage collection from freeing up memory.

Solution: Ensure you explicitly clear large variables or objects as soon as they are no longer necessary. In the context of GUI development, this might mean clearing temporary data after it has been displayed or processed. Utilize MATLAB's “guidata” function to manage GUI data more effectively, ensuring that only necessary data is kept alive.

3 Inefficient GUI Updates and Callbacks

Issue: Frequent or unnecessary updates to the GUI elements can lead to excessive memory usage, especially if these updates involve large data sets.

Solution: Minimize the frequency of GUI updates and make sure that data passed to GUI elements for display purposes is appropriately sized. For instance, if you're displaying images or plots, consider downsampling or compressing the data before displaying it. Implement logic within your callbacks to prevent unnecessary updates or computations.

4. Suboptimal Data Handling and Structure

Issue: Passing large data structures between functions, especially within a GUI, can quickly consume memory, particularly if copies of the data are made.

Solution: Where possible, pass references to large data structures instead of copying them. Consider restructuring your code to minimize the scope of large variables, using them locally within functions to ensure they are cleared from memory when the function execution completes. For global access, explore using MATLAB's `persistent` variables or storing data in a centralized structure that can be accessed by reference.

You can also refer to the below document for a clearer understanding of the functions and workaround mentioned in the above script.

https://in.mathworks.com/help/matlab/matlab_prog/resolving-out-of-memory-errors.html

https://in.mathworks.com/help/matlab/ref/guidata.html

Hope this helps!