The very mention of Artificial Intelligence (AI) brings to mind images of an intelligent world. Well, let’s just say we are heading in the right direction – because the capabilities showcased by this new-age technology are endless.
In recent research conducted in Germany, it was revealed that AI can correctly detect interval breast cancers, which can often develop between screening rounds. These often go undetected by radiologists, or come up as a false positive result. The study was conducted by a tech-startup Vara, in partnership with The Mammography Reference Centre North in Oldenburg. It was later published in the European Journal of Radiology.
This is a positive piece of news to deal with the increasing occurrence of breast cancer, over the years. According to the World Health Organisation (WHO), there were 2.3 million women who were diagnosed with breast cancer. Furthermore, there were 685,000 deaths recorded globally. That’s not all – it is the most commonly diagnosed cancer in Indian women.
The high mortality in breast cancer
Breast cancer is associated with high mortality rates, particularly in India. Why is that so? Because the critical illness is not detected in time. There are also certain other facts that come into play – Mammography is said to have a sensitivity of 62-68% in women who have dense breast tissue, as revealed by a study titled ‘Individual and Combined Effects of Age, Breast Density, and Hormone Replacement Therapy Use on the Accuracy of Screening Mammography’, published in the journal Annals of Internal Medicine.
The same applies to breast ultrasonography – the only difference is that the sensitivity is more pronounced in younger women with dense breast tissue. In such cases, clinical breast examination is the preferred choice and far more affordable, since it does not require any equipment. This is solely dependent on the medical practitioner, but due to the increasing healthcare burden, there may be certain misses. This is alarming, because not being detected properly can claim lives.
In such situations, it is only wise to choose tech-heavy solutions that are both accessible and affordable, and improve patient outcomes. This is where AI presents itself as a saviour.
Affordable and reliable solutions to detect breast cancer
So, what is the best way to use AI in the early diagnosis of breast cancer? By combining medical imaging with AI, it is possible to accurately detect illness, and in a quick manner.
According to a study titled ‘External Evaluation of 3 Commercial Artificial Intelligence Algorithms for Independent Assessment of Screening Mammograms’ published by JAMA Oncology, “researchers have worked hard to leverage the capabilities of AI, or more specifically of deep neural networks, to enable computer-aided detection (CAD) for independent mammography assessments. The reported performance levels have in several cases been on par with radiologists.”
With AI, machine learning, and deep learning, automation of assessments can happen and a radiologist simply needs to review the outcome. Moreover, these technologies can see what humans can’t and that’s where their potential really lies.
Niramai’s Thermalytix is one of the most successful AI-integrated solutions for breast cancer. It is a computer-aided diagnostic engine that is powered by AI. It makes use of a high-resolution thermal sensing device, and a cloud-hosted analytics solution for analyzing the thermal images. The good news is that this can also be used by a non-trained clinician, who has basic tech skills. What this essentially means is that the radiologist doesn’t have to be particularly present to carry out the diagnosis.
There’s another group that has created an AI algorithm that can advise people on the duration of their screenings – how often they should go for it. This model makes use of their mammogram images to evaluate their risk of developing breast cancer, in the next 5 years or so.
The last word
What’s important to understand is that using AI solely for the diagnosis will not serve well. Instead, it must be closely integrated into the entire clinical routine. This way, both radiologists and AI have all the feedback to understand how the process is progressing. Currently, this feedback can take up to several months.
The path towards unleashing the potential of AI for breast cancer diagnosis has been carved – it’s only likely to strengthen over a period of time!
Source: indiaai.gov.in
