In a significant advancement at the crossroads of artificial intelligence (AI) and healthcare, Tobi Titus Oyekanmi, a computer scientist from New Mexico Highlands University, has introduced a deep learning model poised to change the landscape of brain cancer diagnosis, particularly in areas with limited resources. His research, titled “Deep Learning-Based Diagnosis of Brain Cancer Using Convolutional Neural Networks on MRI Scans: A Comparative Study of Model Architectures and Tumor Classification Accuracy,” features a new system called LightBT-CNN, which attained an impressive 98% diagnostic accuracy on MRI scans of the brain.
The findings were published in the American Academic Scientific Research Journal for Engineering, Technology, and Sciences (ASRJETS), positioning Oyekanmi as a leading figure in the burgeoning field of explainable artificial intelligence (XAI) for medical imaging. Brain cancer, recognized as one of the most lethal cancer types globally, typically relies on manual interpretation of MRI scans, a method that can be laborious and subject to mistakes.
Oyekanmi stated, “AI can help level the playing field. The goal was to build a lightweight yet powerful neural network that can analyze brain MRI scans with accuracy comparable to expert radiologists but without the need for expensive infrastructure.” To realize this vision, he led a diverse team that included Peter Adigun, Nelson Azeez, and Ayodeji Adeniyi. They developed the LightBT-CNN, a convolutional neural network capable of classifying four types of tumors: glioma, meningioma, pituitary, and healthy brain scans, utilizing a dataset of over 7,000 MRI images.
In contrast to large deep learning models such as VGG16 or ResNet50 that demand high-performance GPUs, Oyekanmi”s LightBT-CNN is designed with only 3.6 million trainable parameters, making it compact, cost-efficient, and suitable for healthcare facilities in developing nations. Constructed using Python and TensorFlow, the model demonstrated exceptional precision and recall rates exceeding 95% across all tumor categories.
A key feature of this system is its ability to provide interpretability. Through Gradient-weighted Class Activation Mapping (Grad-CAM), the model visually indicates which areas of the brain influenced its predictions, offering physicians clear insights into the diagnosis. Oyekanmi remarked, “Trust is everything in medicine. If an AI can show why it made a decision, clinicians are more likely to adopt it.”
Although Oyekanmi is based in the United States, he maintains strong ties to Nigerian research networks, collaborating with physicist Nelson Abimbola Azeez from the University of Abuja. Together, they aspire to leverage AI to tackle diagnostic challenges throughout Africa, addressing issues ranging from brain tumors to pneumonia detection. Oyekanmi emphasized, “This is not just about publishing papers. It”s about creating practical tools that improve patient outcomes and build local capacity in medical AI.”
His previous collaborations with Adigun and Adeniyi on AI-driven X-ray interpretation for pneumonia detection laid the groundwork for this extensive brain cancer initiative, illustrating a consistent commitment to harnessing AI for societal benefit. Earlier this year, Oyekanmi received the 2025 NIPES Award for Outstanding Contribution to Research and Innovation from the National Institute of Professional Engineers and Scientists (NIPES). He was selected from over 1,200 nominations across four countries, with recognition given for his innovative work, measurable impact, and adherence to international research standards.
Oyekanmi expressed the significance of this recognition, stating, “Recognition from NIPES means a lot to me. It reminds me that impactful research isn”t just about algorithms; it”s about improving lives.” Experts have praised Oyekanmi”s efforts for merging academic rigor with practical applications. The study evaluates LightBT-CNN”s performance against global benchmarks like ResNet and EfficientNet, demonstrating comparable accuracy while requiring significantly less computational power, a critical factor for healthcare systems with limited digital infrastructure.
“AI doesn”t have to be complicated to be effective. Sometimes simplicity and efficiency matter more than brute-force computation,” Oyekanmi added. Despite these achievements, he acknowledged the limitations of the study, noting that the dataset reflects controlled MRI conditions and does not capture the variability often present in real-world hospital environments.
Looking forward, Oyekanmi intends to collaborate with clinical partners to validate the system using actual patient data and investigate multi-modal imaging approaches that incorporate MRI with CT and PET scans. He also advocates for cross-institutional AI training programs at Nigerian universities to equip the next generation of scientists with practical experience in medical machine learning. “This work reminds us that innovation isn”t confined to big tech companies. With the right vision, collaboration, and compassion, AI can become a tool for equity, helping every patient, everywhere, get the care they deserve,” he reflected.
