Professor Pallavi Tiwari’s AI Journey
Dr. Pallavi Tiwari, a professor of radiology and biomedical engineering at the University of Wisconsin-Madison, has spent 18 years advancing artificial intelligence (AI) models to revolutionize cancer research. Her work focuses on leveraging machine learning to predict cancer diagnoses, patient outcomes, and drug responses.
“We’ve been using AI approaches to identify overall patient outcomes and their response to specific drugs,” Tiwari explained.
Joining UW-Madison in 2022, Tiwari leads a campus-wide AI initiative aimed at enhancing healthcare and medical imaging.
Glioblastoma Study: Insights into Sex-Based Tumor Differences
Earlier this year, Tiwari and her research team investigated why glioblastoma—a lethal brain cancer—is more aggressive in men. Their findings, published in Science Advances, highlighted biological differences in how tumors develop in men versus women.
Using AI, the researchers analyzed data from 250 glioblastoma studies across various institutions to ensure robust and tumor-specific insights. The AI model identified common and distinct characteristics in tumors based on the patient’s sex.
For women, higher-risk tumors tended to invade healthy tissue, while in men, the presence of specific cells near dying tissue was associated with more aggressive tumor behavior.
“These findings could guide personalized treatments, identifying the right candidates for therapies like immunotherapy or chemotherapy,” Tiwari noted.
AI as a Medical Ally
Tiwari believes AI holds transformative potential for healthcare, particularly in managing vast datasets beyond human capability. AI could analyze thousands of treatment data points to determine effective therapies and their influencing factors, offering a valuable decision-support tool for oncologists.
“Ultimately, AI becomes an ally, helping doctors make better decisions and ensuring patients receive the most appropriate treatment,” Tiwari emphasized.
This groundbreaking research not only advances cancer diagnostics but also paves the way for more personalized and effective treatment strategies.
News 2: FDA Grants Priority Review to Verastem Oncology’s NDA for Recurrent Low-Grade Serous Ovarian Cancer Treatment
NDA for Avutometinib and Defactinib Combination Accepted
Verastem Oncology, a biopharmaceutical company focused on innovative cancer treatments, announced that the U.S. Food and Drug Administration (FDA) has accepted its New Drug Application (NDA) under the accelerated approval pathway. The NDA covers the combination of avutometinib, an oral RAF/MEK clamp, and defactinib, an oral FAK inhibitor, for treating adults with recurrent low-grade serous ovarian cancer (LGSOC) carrying KRAS mutations. Patients eligible for this treatment must have undergone at least one prior systemic therapy. The FDA granted Priority Review, setting a Prescription Drug User Fee Act (PDUFA) action date of June 30, 2025, and indicated no plans for an advisory committee meeting.
Significance of the Approval
LGSOC is a rare and persistent ovarian cancer with no FDA-approved treatments currently available. Verastem’s combination therapy aims to address this significant unmet medical need. Dan Paterson, CEO of Verastem Oncology, expressed enthusiasm about the potential to deliver the first FDA-approved treatment for recurrent KRAS mutant LGSOC in the U.S., emphasizing the importance of this milestone for patients.
Clinical Evidence and Ongoing Trials
The NDA is supported by data from the Phase 2 RAMP 201 trial, which demonstrated substantial and durable responses in patients. The combination therapy was also well-tolerated. Additional supportive data comes from the FRAME Phase 1 trial. Verastem is conducting the Phase 3 RAMP 301 trial to further evaluate the treatment’s efficacy, which could expand its indication beyond KRAS mutant cases.
Addressing a Rare Cancer
LGSOC primarily affects younger women, often having a profound impact on quality of life. Current treatments include chemotherapy and hormone therapy, but their efficacy is limited. Verastem’s investigational therapy represents a significant advancement for this patient population.
Verastem Oncology remains committed to delivering innovative treatments, with a commercial launch expected in mid-2025, pending FDA approval.
News 3: Study Reveals Human Brain’s Surprisingly Slow Thinking Speed Compared to Technology
Slow Thinking Compared to Modern Technology
The human brain, with its 86 billion neurons, is a biological masterpiece. However, when compared to modern technology, its processing speed appears surprisingly slow. A recent study published in Neuron reveals that the human brain processes thoughts at a mere 10 bits per second. In contrast, an average Wi-Fi connection in the United States downloads data at over 260 million bits per second, while a simple phone call transmits 64,000 bits per second.
“It’s a counterweight to the hyperbole about the brain’s complexity,” says Markus Meister, study co-author and neuroscientist at the California Institute of Technology. “We are incredibly slow if you put numbers to it.”
Measuring Cognitive Speed
The study calculated the brain’s thinking speed by analyzing human activities, such as typing and video gaming. Advanced typists, who type at 120 words per minute, and elite gamers with lightning-fast reactions both process information at just 10 bits per second. Even in less physically demanding tasks like blindfolded Rubik’s cube solving, cognitive processing caps at 12 bits per second.
Filtering a Flood of Sensory Input
Despite receiving around 10 billion bits of sensory data per second, the brain filters and processes only a tiny fraction. This selective processing raises questions about how the brain manages to focus on so little amidst an overwhelming influx of information.
Evolutionary Design and Modern Challenges
Experts suggest the brain’s slow processing speed may stem from evolutionary constraints. Early humans required only basic navigation and single-path thinking to survive. However, the fast-paced digital world now far outstrips the brain’s capacity.
Neuroscientist Tony Zador remarks, “Nature seems to have built a speed limit into our conscious thoughts.” Despite its limitations in multitasking and parallel processing, the human brain remains a fascinating enigma of evolutionary design.
