Pain has long been described as one of medicine's most personal experiences—real to the person feeling it yet difficult for doctors to measure objectively. Now, researchers in China believe they have taken a meaningful step towards narrowing that gap by developing a new imaging model that looks beyond the brain to include the spinal cord, potentially offering a more accurate way to understand both short-term and chronic pain.

The study, published in Cell Reports Medicine, was led by Dr Kong Yazhuo and colleagues at the Institute of Psychology. Their team combined functional magnetic resonance imaging (fMRI)—a scan that tracks changes linked to nerve activity—with machine learning, a form of computer software trained to recognise patterns in large amounts of data. Rather than relying only on brain signals, the researchers analysed the combined activity of the brain and spinal cord to create what they call the 'Corticospinal Pain Intensity Pattern' (CsPIP).

Pain does not begin in the brain alone. Signals triggered by injury or harmful heat first travel through the spinal cord before reaching the brain, where emotions, memories and thoughts help shape how pain is ultimately experienced. Yet most existing pain biomarkers—measurable biological signals used to estimate disease or symptoms—have focused almost entirely on the brain because imaging both structures at the same time has remained technically difficult.

Using data from healthy volunteers exposed to controlled heat, the researchers trained the model before testing it in independent studies involving heat pain, electrical pain, itch, pain empathy, nerve stimulation and patients living with irritable bowel syndrome, a chronic digestive disorder often accompanied by persistent abdominal pain. The new model consistently predicted participants' reported pain levels more accurately than approaches based on either the brain or spinal cord alone.

Importantly, CsPIP responded to genuine physical pain but not to itching or when volunteers merely watched someone else experience pain, suggesting that it detects first-hand pain rather than emotional responses. The model also reflected pain relief following transcutaneous electrical nerve stimulation and tracked improvements after electroacupuncture in a subgroup of patients with irritable bowel syndrome.

"Our findings suggest that pain perception is better captured by integrated corticospinal activity than by brain activity alone," said Dr Kong.

The findings build on growing evidence that chronic pain involves networks extending beyond the brain. While larger studies will be needed before the technique reaches routine clinical practice, researchers say integrating spinal cord activity could eventually help doctors assess pain more objectively, monitor treatment response and move closer to personalised care for millions living with chronic pain.