University of California, Berkeley - MIDS Capstone Project

Frequently Asked Questions



How do I upload a blood smear image for analysis?
To upload a blood smear image, click the "Upload" button and select the image file from your device. Once selected, click "Submit" to initiate the analysis.

How long does it take to receive analysis results?
The analysis typically takes just a few moments.

What information will I receive in the analysis report?
The analysis report will inform whether the blood smear image is accepted or needs to be retaken or adjusted. If this is the case, a message will describe the problem with the image and how it could be corrected.

If the image is accepted, the report will provide information on whether the image is normal or abnormal.

What should I do if the analysis results indicate that the blood smear may be abnormal?
If the analysis suggests that the blood smear may be abnormal, consulting a medical professional for further evaluation and diagnosis is crucial. The system serves as a screening tool, not a substitute for medical expertise.

Can I submit multiple images at once for analysis?
No, submit each image separately.

Can I save or download the analysis reports for future reference?
Yes, the user will receive a PDF report with summary details on the model, the uploaded image, and the results from the diagnostic tool.

How accurate is the analysis performed by the system's image recognition model?
The accuracy of the analysis depends on the quality and characteristics of the uploaded image, with an accuracy rate of 97.5%. However, it is important to remember that healthcare professionals should make final diagnosis and clinical decisions.

Is patient data collected or retained by the system?
No, we do not collect or retain patient information. Patient privacy is of utmost importance to us. The uploaded images are anonymized and processed solely for analysis. Patient privacy is fully respected and protected.

What is the policy on informed consent for using this diagnostic tool?
Before using our diagnostic tool, obtaining informed consent from the patient or their legal guardian is mandatory. This consent should acknowledge that the patient understands the tool's purpose and process and that the results are for preliminary screening purposes only and not a definitive diagnosis. The consent must also affirm understanding the privacy policy, including how the data will be used and protected. It is important to ensure that the patient or their guardian freely agrees to proceed with the understanding of all these aspects.

Is there any cost associated with using this diagnostic tool?
No, there is no cost. Claudia Durkin, Vincent Gu, and Sam Won, graduate students at the University of California, Berkeley, developed this image recognition model. They hope their work will lead to increased early childhood cancer diagnosis, helping save lives, and there is a World for Every Child.