Explore the potential of AI in breast cancer diagnosis with Meri Sehat. Stay informed for early detection and effective treatment.

Can AI Help Diagnose Breast Cancer?

Accounting for more than 2 million new cases in 2020, breast cancer has now become the most common cancer globally.

Introduction

It is also the most common cancer among women and the leading cause of cancer-related deaths among the gender<a href="#accordion">[1][2]</a>.

In Pakistan, <a href="https://merisehat.pk/disease/breast-cancer">breast cancer </a>was responsible for 14.5% of all new cancer cases in 2020 among both women and men<a href="#accordion">[3]</a>. The country currently has the highest rate of breast cancer in all of Asia, with 1 in 9 women being diagnosed with breast cancer<a href="#accordion">[4]</a>.

Although high income countries have the highest incidence of the disease, the mortality rate is much lower, compared to low- and middle-income countries<a href="#accordion">[5]</a>. This is linked to better screening and treatment facilities in high-income nations. 
&nbsp;

Approximately, 89% of breast cancer patients in Pakistan receive a diagnosis at a later stage, while 59% present at an advanced stage<a href="#accordion">[6]</a>. This delay in diagnosis is linked to reduced survival rates and an increase in mortality rates. The lack of awareness, limited access to healthcare facilities, and lack of effective screening programs all contribute to the delay in diagnosis.

In an effort to reduce the number of deaths, researchers are now turning to artificial intelligence (AI). But can AI play a role in screening for breast cancer? 
&nbsp;

Understanding of AI In Medical Field

Artificial intelligence (AI) is a technology that enables computers to perform tasks that typically require human intelligence. It involves creating computer programs and algorithms that can learn from data, recognize patterns, and make decisions or predictions.

Doctors and researchers use AI algorithms to analyze medical images, such as mammograms, MRIs, and ultrasounds, to detect potential signs of breast cancer. These algorithms can identify subtle patterns or abnormalities that may be difficult for human eyes to notice. By helping doctors with image analysis, AI improves the accuracy of breast cancer diagnosis and reduces the chances of missed or delayed diagnoses<a href="#accordion">[7]</a>.

AI is also used to predict an individual&#39;s risk of developing breast cancer. Furthermore, AI models can provide personalized risk assessments by analyzing factors, like personal health records, genetic information, lifestyle factors, and mammogram results. These assessments enable doctors to identify individuals at higher risk and implement appropriate preventive measures or more targeted screening programs.

&nbsp;

Mammograms are the universal screening test for breast cancer. However, mammograms can miss about 1 in 8 cases of breast cancer<a href="#accordion">[14]</a>. The incorporation of AI can help reduce false-negative values and increase cancer detection.

A recent population-based cohort study compared AI reading of digital mammograms with radiologists&rsquo; reading. The researchers aimed to compare the accuracy and the difference in cancer detection by AI and the standard human double-reading practice<a href="#accordion">[10]</a>. 
&nbsp;

According to the results, AI was able to detect interval cases that radiologists did not detect. The use of AI also resulted in lower overall screen-reading volume. Thus, indicating the potential of AI as a screen-reader of mammograms, and reducing workload for radiologists.

Another study combined the strengths of AI and radiologists for breast cancer screening. As per the results, the combined system detected 2.6% more breast cancer with fewer false alarms<a href="#accordion">[12]</a>. 
&nbsp;
<quillbot-extension-portal></quillbot-extension-portal>

A Norwegian population-based study compared the performance of AI with standard double reading. They assessed more than 100,000 examinations from more than 47,000 women across Norway<a href="#accordion">[10]</a>.

A scale of 1 to 10 was used by the AI system to predict the risk of cancer, with 10 representing the highest risk. The AI system gave a score of 10 to 86.8% of screen-detected cancers and 44.9% of interval cancers. Less than 20% of screen-detected cancers were not selected by the AI system. Thus, AI proved to be a promising tool for detecting breast cancer.

Another study evaluated the use of a deep learning algorithm for predicting the risk of breast cancer based on digital mammograms<a href="#accordion">[11]</a>. The AI system was found to be a stronger predictor of breast cancer risk following a negative mammographic screening for 3-6 years. 
&nbsp;

CAD systems, powered by AI, can assist radiologists in interpreting mammograms. These systems analyze mammographic images, highlighting potential areas of concern for further evaluation. CAD helps radiologists in identifying suspicious lesions or calcifications that might be indicative of breast cancer, enhancing the overall accuracy of the diagnostic process.

In a recent study, researchers used CAD to detect microcalcifications and classify mammograms<a href="#accordion">[13]</a>. More than 4,000 mammograms with 6,663 microcalcifications were assessed using the system. The AI system accurately detected and classified microcalcification lesions as benign or malignant. The overall classification accuracy for distinguishing between benign and malignant breasts was 72.4%, with a sensitivity of 77.8%. At the image level, the overall accuracy was 70.6%, and at the microcalcification lesion level, it was 75.1%.

In conclusion, the study demonstrated that an AI system focusing on microcalcifications can achieve accurate early diagnosis of breast cancer. However, further research is needed to improve the classification of pathological types and molecular-level factors such as ER status. 
&nbsp;

Conclusion

Overall, AI in breast cancer diagnosis helps doctors make more accurate diagnoses by analyzing medical images and predicting the risk of developing the disease. It acts as a valuable tool that complements doctors&#39; expertise, improving patient outcomes and enhancing the effectiveness of breast cancer screening and prevention efforts. 
&nbsp;

However, it&#39;s important to note that while AI holds immense potential in the field of breast cancer, it is not meant to replace medical professionals. Rather, it serves as a powerful tool that augments their capabilities, improves decision-making, and enhances patient care. Continued research, rigorous validation, and integration of AI into clinical practice will be crucial for harnessing its full potential and ensuring its safe and effective implementation.

[1] Arnold M, Morgan E, Rumgay H, et al. Current and future burden of breast cancer: Global Statistics for 2020 and 2040. The Breast. 2022;66:15-23. doi:10.1016/j.breast.2022.08.010

[2] Pavani Chalasani M. Breast cancer. Practice Essentials, Background, Anatomy.

[3] International Agency for Research on Cancer, 2021. Pakistan Globocan 2020.

[4] Khan NH, Duan SF, Wu DD, Ji XY. Better Reporting and Awareness Campaigns Needed for Breast Cancer in Pakistani Women. Cancer Manag Res. 2021;13:2125-2129. Published 2021 Mar 2. doi:10.2147/CMAR.S270671

[5] Saeed, S., Asim, M. & Sohail, M.M. Fears, and barriers: problems in breast cancer diagnosis and treatment in Pakistan. BMC Women's Health 21, 151 (2021)

[6] Gulzar, F. et al., 2019. Identifying the reasons for delayed presentation of Pakistani breast cancer patients at a tertiary care hospital. Cancer Management and Research, Volume 11, pp. 1087-1096

[7] Dileep G, Gianchandani Gyani SG. Artificial Intelligence in Breast Cancer Screening and Diagnosis. Cureus. 2022;14(10):e30318. Published 2022 Oct 15. doi:10.7759/cureus.30318

[8] Marinovich ML, Wylie E, Lotter W, et al. Artificial Intelligence (AI) for breast cancer screening: BreastScreen population-based Cohort Study of Cancer Detection. eBioMedicine. 2023;90:104498.

[9] Lehman CD, Arao RF, Sprague BL, et al. National Performance Benchmarks for Modern Screening Digital Mammography: Update from the Breast Cancer Surveillance Consortium. Radiology. 2017;283(1):49-58. doi:10.1148/radiol.2016161174

[10] Larsen M, Aglen CF, Lee CI, et al. Artificial Intelligence Evaluation of 122969 mammography examinations from a population-based screening program. Radiology. 2022;303(3):502-511. doi:10.1148/radiol.212381

[11] Damiani C, Kalliatakis G, Sreenivas M, et al. Evaluation of an AI model to assess future Breast Cancer Risk. Radiology. 2023;307(5). doi:10.1148/radiol.222679

[12] Leibig C, Brehmer M, Bunk S, Byng D, Pinker K, Umutlu L. Combining the strengths of Radiologists and AI for Breast Cancer Screening: A retrospective analysis. The Lancet Digital Health. 2022;4(7). doi:10.1016/s2589-7500(22)00070-x

[13] Lin Q, Tan W-M, Ge J-Y, et al. Artificial intelligence-based diagnosis of breast cancer by mammography microcalcification. Fundamental Research. Published online 2023. doi:10.1016/j.fmre.2023.04.018

Trusted By