September is Blood Cancer Awareness Month, a time to raise awareness of this devastating disease and its impact on pediatric patients. Blood cancers are the most common type of cancer in children and can be tough to treat. Many children with blood cancer are forced to undergo aggressive treatments that can have lasting effects on their physical and emotional health. As discussed in our last blog dedicated to blood cancer awareness, pediatric patients may experience side effects such as fatigue, nausea, and hair loss. They may also have to miss school and social activities. Despite the challenges, many children with blood cancer live long and healthy lives. With the support of their families, friends, medical teams, and new technologies, they can hopefully overcome this disease and thrive! 

In this installment dedicated to blood cancer awareness, we’ll cover how Artificial intelligence (AI) has been used to develop new diagnostic tools for blood cancers in recent years. Some people worry that AI will become so powerful that humans will lose control. This is a legitimate concern, as AI systems are becoming increasingly complex and sophisticated. However, it is essential to remember that AI systems are still designed and created by humans. Humans have the power to control how AI systems are used.   

How Can AI Help? 

Artificial intelligence (AI) is being used to diagnose blood cancers in many ways, including: 

• Automated image analysis: AI can analyze blood cell images to identify abnormalities indicative of cancer. This can be used to screen blood samples for cancer or to help doctors diagnose cancer more accurately. 
• Machine learning: Machine learning algorithms can be trained on large datasets of blood cancer images to learn to identify patterns associated with different cancer types. This can be used to develop new diagnostic tools that are more accurate than traditional methods. 
• Natural language processing: Natural language processing (NLP) can be used to analyze patient medical records to identify risk factors for blood cancer or to track the progression of the disease. This information can be used to help doctors make better treatment decisions. 

• Chatbots: Chatbots can provide information and support to patients with blood cancer. They can also be used to collect data from patients that can be used to improve AI-powered diagnostic tools. 

Current Research on AL 

If you have a child with cancer, specifically blood cancer, you likely already know that the most common blood cancer in children is acute lymphoblastic leukemia (ALL). It accounts for about 30% of all childhood cancers. The good news is that AI offers hope for earlier detection of ALL! Learn more about the research conducted to use AI as a tool or companion diagnostic system to predict ALL earlier in potential cancer patients. Check out the studies below! 

• Predicting Molecular Subtype and Survival of Rhabdomyosarcoma Patients Using Deep Learning of H&E Images: A Report from the Children’s Oncology Group-This is the first study to include a wider variety of cell types in leukemia diagnosis, and the results suggest that the system could be a valuable tool for improving the early detection and treatment of leukemia in children. A new was developed that uses a convolutional neural network (CNN) to analyze bone marrow images and was able to achieve an accuracy of 82.93% in classifying white blood cells (WBCs) and 89% in diagnosing acute lymphoid leukemia (ALL). The system also detected the bone marrow metastasis of lymphoma and neuroblastoma. The study also found that the AI system could identify high-risk features associated with specific mutations at diagnosis. This is important because it could help doctors better predict the disease’s course and develop more personalized treatment plans. 

• Explainable AI in Diagnosing and Anticipating Leukemia Using Transfer Learning Method-This study proposed an automated method for detecting acute lymphoblastic leukemia (ALL) using transfer learning models and local interpretable model-agnostic explanations (LIME). LIME is a technique for explaining the predictions of black-box machine learning models. A black-box model is a model that cannot be easily understood by humans. LIME works by creating a more straightforward, interpretable model that explains the predictions of the black-box model. The method was evaluated on a dataset of blood cell images and achieved an accuracy of 98.38% with the InceptionV3 model (in lamins terms, that’s a tool for image recognition.) The results showed that the proposed method is a promising approach for the early detection of ALL. 

• A2M-LEUK: attention-augmented algorithm for blood cancer detection in children-A new algorithm called A2M-LEUK has been proposed for the early detection of leukemia. A2M-LEUK uses attention mechanisms to focus on the most essential features in blood cell images, making it more sensitive to subtle changes that can indicate leukemia. Attention mechanism basically means it allows the model to “pay attention” to certain parts of the data and to give them more weight when making predictions. The algorithm was more accurate than other methods, achieving an accuracy of 98.38% on a dataset of blood cell images. This optimistic development could lead to earlier detection and treatment of leukemia, improving patient outcomes.  

Need Cancer Support? 

Here To Serve can support your family as you navigate your child’s cancer diagnosis. As the only national nonprofit focusing on the cancer journey, helping families with their day-to-day life, including finances, meals, housekeeping, childcare, pet care, transportation, and more, we are the missing link you might so critically need. Please don’t hesitate to contact us and get help immediately! 

References: 

• https://www.explorationpub.com/Journals/etat/Article/1002127 
• https://www.ibm.com/topics/artificial-intelligence 
• https://www.clinicallab.com/standardized-fecal-self-collection-device-helps-microbiome-labs-27254 
• https://www.hindawi.com/journals/cin/2022/5140148/ 
• https://doi.org/10.1007/s00521-023-08678-8 

By Sameera Rangwala, M.S., M.P.H 

About the Author 

Sameera Rangwala spent 15 years in the biotechnology industry and is currently a life science educator for children in grades 5-8.  As a scientist and research professional, she uses her skills to blog and provides words of support to the cancer community. 

All content in this blog is for informational and educational purposes only. Always consult a medical provider in your particular area of need before making significant changes in your medical decisions or lifestyle.