As of: January 18, 2024, 3:59 p.m
By: Juliane Gutmann
Comments
Press
Split
If skin cancer is detected early, the chances of recovery increase immensely.
A study has shown that artificial intelligence is extremely helpful in terms of diagnosis.
Artificial intelligence (AI) makes life easier in many areas.
This is how AI technologies increase the safety of cars and public transport.
In customer service it is used in the form of chatbots and virtual assistants for quick help.
In agriculture, AI optimizes the cultivation and harvesting of crops.
And it also has advantages in medicine.
Here, artificial intelligence supports, for example, the diagnosis of diseases.
In early skin cancer detection, AI software even achieves a hit rate of 100 percent for malignant skin cancer.
The organizers of the congress of the European Academy of Dermatology and Venereology (EADV), which took place in October 2023, will provide information about this.
As part of the study presented at the congress, 22,356 patients with suspected skin cancer were examined over a period of 2.5 years.
The study leaders used AI-supported software to detect cases of skin cancer.
59 out of 59 skin cancer cases detected: AI is convincing in skin cancer diagnosis
The positive result: 100 percent of melanomas – the most serious form of skin cancer – were detected by the software.
This corresponded to 59 out of 59 cases.
But there are also other types of skin cancer.
Here the hit rate was slightly lower: 99.5 percent (189 of 190 cases) were identified as skin cancer by the AI.
A dermatologist must decide whether a mole needs to be removed or not.
Artificial intelligence is helpful when it comes to diagnosing skin cancer.
© Imago
Don't miss anything: You can find everything about health in the regular newsletter from our partner 24vita.de.
“This study has shown how AI is rapidly improving and learning, with high accuracy directly attributable to improvements in AI training techniques and the quality of data used to train AI.
The latest version of the software saved over 1,000 face-to-face visits in secondary care between April 2022 and January 2023. This resulted in more time being available for patients in need of urgent care,” said the study’s lead author, Dr.
Kashini Andrew, Consultant General Practitioner at University Hospitals Birmingham NHS Foundation Trust, quoted on the pages of eadvcongress2023.de.
AI software is not used in Germany
Although the data is very encouraging, the research team cautions that AI should not be used as a sole detection tool without the support of a dermatologist.
“We would like to emphasize that AI should not be used as the sole tool for skin cancer detection and that AI is not a replacement for consulting dermatologists,” says co-author Dr.
Irshad Zaki, consultant dermatologist at University Hospitals Birmingham NHS Foundation Trust.
My news
Prevent biliary cancer: Diabetes and smoking as well as tobacco smoke significantly increase the risk read
No appointment in the practice necessary: Read which specialists you can see directly without a referral
Slim overnight?
Certain types of tea are said to reduce belly fat while you sleep
Detoxification patches for feet: Detox miracle cure or rather a rip-off?read
According to experts, up to five kilos of weight loss per week should be possible - with intermittent fasting
Vitamin D preparations at Ökotest – only five out of 20 products are convincing
However, the researchers add that further research under appropriate clinical supervision could enable the use of AI as a triage tool.
The researchers see great potential for future healthcare.
The AI software examined is currently not used in standard care for the early detection of skin cancer in Germany.
You can read here how to recognize skin cancer using the ABCDE list.
This article only contains general information on the respective health topic and is therefore not intended for self-diagnosis, treatment or medication. It in no way replaces a visit to the doctor. Unfortunately, our editors are not allowed to answer individual questions about medical conditions.