SECTIONS
Sunday, February 24, 2019
CEBU

Early skin disease detection through apps

SKIN conditions can sometimes be ignored to our peril. Allergic red rashes on the skin, for instance, may appear subtly. However, its development in the next few minutes can be critical to the person’s life. Once, when I was a human resource officer in a Talamban-based group of companies, I had to drive a division manager to a nearby hospital due to a sudden surge of red rashes on her skin. Cases like this are normally treated in hospitals as emergency cases. Each second is crucial as it is life-threatening.

Skin diseases can be caused by three major groups of micro-structures: bacteria (living), fungi (living), viruses (living), genes (living) and allergens (non-living). Thus, skin diseases can be very numerous to memorize unless you want to. They can range from rashes and other allergies to infectious forms, such as acne (caused by a species of bacteria), ringworm (caused by a species of fungi), shingles (caused by a species of viruses) and cancer lesions.

Internet-based skin detection systems are now available but may not be physically in our country. Most of these systems work using image processing and data mining technologies, which are capable of pre-processing and segmenting scanned images of the suspected skin area, and those clinically relevant skin features are then extracted electronically and compared to millions of images of diverse skin diseases in the provider’s archive.

Using a downloadable app, Windows-based tabs can be used to capture the skin area image and upload it into a skin detection portal. You will get results in a few minutes or less.

The image processing technology has been known to have different accuracies across different skin diseases. Nisha Yadav, Virender Narang and Utpal Shrivastava reported in the March 2016 issue of the International Journal of Computer Applications that it is, for instance, 85 percent accurate in recognizing eczema and melanoma and 95 percent accurate in recognizing impetigo.

Meanwhile, data mining technology is now capable of identifying the emerging skin disease and recommendations to prevent its further development or its treatment.

These online systems are fast, highly accurate across all skin types, and can quickly classify skin diseases and recommend preventive and treatment measures. There are a few disadvantages, though. Since this application of image processing and data mining is still new, some skin diseases are still not fully studied for accuracy verification.

Right now, most, if not all, of these apps run only in Windows applications. Thus, iOS and Android users may find difficulty finding online providers that use compatible apps. Perhaps, early skin disease detection may still be difficult for Filipinos, after all.


VIEW COMMENTS
DISCLAIMER:

SunStar website welcomes friendly debate, but comments posted on this site do not necessarily reflect the views of the SunStar management and its affiliates. SunStar reserves the right to delete, reproduce or modify comments posted here without notice. Posts that are inappropriate will automatically be deleted.


Forum rules:

Do not use obscenity. Some words have been banned. Stick to the topic. Do not veer away from the discussion. Be coherent. Do not shout or use CAPITAL LETTERS!