Noman Khan
A few conditions strike as suddenly and devastate as completely as a subarachnoid hemorrhage (SAH), a type of bleeding around the brain that can kill within hours if not treated.
Across the world, SAH remains a leading cause of sudden neurological death and long-term disability. In Pakistan, where access to neurosurgical care varies sharply between regions, the burden is particularly heavy. Many patients arrive late to specialized centers, while families and doctors struggle to predict outcomes with limited tools.
It is in this context that Dr Muhammad Mohsin Khan, a neurosurgeon and neuroscientist working in a leading clinical and research institute in Qatar, has stepped forward with groundbreaking work. His latest study, published in Neurosurgical Review, introduces a machine-learning model that predicts SAH severity with remarkable accuracy. In an interview, Dr Khan explained that this research is not just about algorithms but about saving lives and guiding scarce resources in countries like Pakistan.
“When a patient with SAH is admitted, doctors and families want to know the chances of survival. These are heartbreaking conversations and until now, our prediction tools have been blunt instruments,” he said. His model, using a three-stage approach, classifies patients into progressively finer outcome categories, addressing the long-standing issue of rare but critical outcomes being overlooked by older systems.
The logic is straightforward: patients are first separated into good or poor outcome groups, then further classified using refined predictors. The result is a more precise forecast of recovery or disability, supported by nearly ninety percent accuracy. “We designed this model to be practical. It uses data any hospital can collect at admission. That means it can work in Karachi, Lahore or Peshawar,” Dr Khan emphasized.
The implications for Pakistan are enormous. SAH strikes mostly middle-aged adults, often breadwinners, leading to economic and social hardship. In a country with limited health budgets, accurate prognostic tools could help hospitals allocate intensive care beds, prioritize surgical interventions and communicate with families more responsibly. “Imagine a district hospital being able to triage who urgently needs transfer to a neurosurgical unit and who can be managed locally. That is the kind of practical impact we are aiming for,” he explained. Globally, SAH accounts for about five percent of all strokes but has a disproportionately high mortality rate. In Pakistan, underreporting hides the true scale, but local data suggest outcomes are often worse than in high-income countries due to delayed diagnosis and treatment. By introducing a validated model, Dr Khan’s research offers a way to bridge these gaps. Hospitals can integrate the tool into electronic health systems or even use it manually to guide early decisions.
“The human side of this work is never far from my mind,” Dr Khan reflected. “A tool that can give families and doctors early clarity is more than an academic contribution—it is humane.” Looking ahead, he plans to expand validation through multicentre studies and adapt the system for use in resource-limited hospitals. “Our vision is global, but the heart of this research beats for patients in places like Pakistan, where the difference between life and death often comes down to rapid, informed decisions,” he concluded.
—The writer, based in Islamabad, occasionally contributes to the national press.
