Mental health treatments such as transcranial magnetic stimulation (TMS) have helped many individuals with major depressive disorder. In standard practice, TMS protocols are typically applied uniformly, with an electromagnetic coil used to deliver magnetic pulses to the same predefined brain region across all patients.

The National University Hospital (NUH) and the Yong Loo Lin School of Medicine, National University of Singapore (NUS Medicine) are exploring a more personalised approach through two frontier studies. By combining advanced brain imaging with tailored TMS targeting, the research aims to identify and treat the specific brain circuits linked to each patient's symptoms.

The first, a now-completed study in patients with severe obsessive-compulsive disorder (OCD), is described as one of the first in the world to guide this type of TMS using “causal” brain circuits – circuits known to relieve OCD symptoms when stimulated through brain surgery – and has produced encouraging early results. The second, in patients with depression and anxiety, begins in October 2026.

While research is still evolving, past studies suggest that this strategy may lead to improved outcomes. A clinical study on standard TMS showed a 21 per cent response rate among participants. In contrast, a separate study on personalised TMS (pTMS) reported far higher success: 70 per cent responded within the month following treatment, with 31 per cent of participants maintaining response after 12 weeks.

The studies described involve participants across distinct clinical groups: one with OCD and one focused on individuals with depression who also experience anxiety symptoms. The studies are led by Adjunct Associate Professor Tor Phern Chern, Senior Consultant, Department of Psychological Medicine, NUH, and Associate Professor Thomas Yeo, Deputy Head, Centre for Translational Magnetic Resonance Research (TMR), NUS Medicine. Assoc Prof Yeo also has a joint appointment with the Department of Electrical and Computer Engineering in the College of Design and Engineering at NUS.

Tailoring TMS to the individual symptoms and brain

Across the studies, participants undergo a standardised pTMS workflow. Functional magnetic resonance imaging (fMRI) is first used to map each participant's unique brain circuitry at the NUS Medicine Centre for TMR. These scans are then analysed using an algorithm first developed by Assoc Prof Thomas Yeo and Dr Ru Kong, who are both also from the Centre for Sleep and Cognition at NUS Medicine.

The resulting data personalises the delivery of magnetic pulses to each individual's identified stimulation target via a neuro-navigation robotic arm, using intermittent theta-burst stimulation (iTBS), a shorter form of TMS in which each treatment session takes about nine minutes rather than the conventional 37.5 minutes. Treatment is administered at the Centre for Interventional Psychiatry at NUH.

To better understand treatment response, participants in these studies undergo intensive monitoring with frequent and repeated fMRI scans across the study period. This more advanced approach, not typically used in routine TMS studies, enables researchers to track how brain circuits change over the course of treatment and how these changes relate to symptom improvement, helping to guide the development of more targeted and personalised interventions.

OCD study shows promise of personalised TMS for severe OCD

This completed study, one of the first in the world to use causal-circuit, fMRI-guided iTBS for OCD, investigated the use of pTMS in four participants with severe, treatment-resistant OCD. OCD is one of the hardest mental health conditions to treat. The only medicines approved for it belong to a single drug class, serotonin reuptake inhibitors, and 40 to 60 per cent of patients do not respond adequately to them.

For those who have not improved with medication and psychotherapy, the most intensive remaining option is deep brain stimulation, which requires brain surgery. A non-personalised form of TMS, cleared for OCD in the United States since 2018, helps about 40 per cent of these patients, but it stimulates the same brain region in every patient.

The OCD research study was supported by a NUH Clinician Scientist Programme grant, and by funding from the Centre for TMR at NUS Medicine. It first uses causal brain circuits – circuits in which Deep Brain Stimulation (DBS) has been shown to relieve OCD symptoms – to identify a general treatment region, which is then refined to each participant's own default mode network using fMRI and the NUS Medicine-developed algorithm to determine each participant's precise stimulation target. Unlike DBS, pTMS is non-invasive and requires no surgery.

Of the four participants enrolled between January and July 2026, two withdrew from the study before completing treatment, while the remaining two completed the full treatment course of 40 short iTBS sessions delivered over five days. Both experienced a substantial reduction in their OCD symptoms, measured using the Obsessive-Compulsive Inventory–Revised (OCI-R), a standard OCD symptom questionnaire.

In one patient, symptom scores fell by 78 per cent by the end of the treatment week and reached zero, with no symptoms reported on the questionnaire, six weeks later.

The other patient improved more gradually, with a 35 per cent reduction in symptoms four weeks after treatment. While this second patient experienced a return of symptoms about 12 weeks later and underwent an additional two-day booster treatment, the patient again showed substantial improvement afterwards. These early findings support further evaluation of pTMS as a safe and effective option for treating OCD. If confirmed in larger studies, it could offer new hope to Singaporeans living with OCD who have not responded well to conventional treatments.

Adj Assoc Prof Tor Phern Chern said, "For people with severe OCD who have tried medication and therapy without success, there have been very few genuinely new treatment options in decades. Being able to reach the circuits that deep brain stimulation targets, without surgery and in a single week of treatment, is what makes these early results so encouraging."

Extending pTMS to depression with co-occurring anxiety symptoms

The other study will involve a larger participant group over a longer study period, examining how brain networks and symptoms change during pTMS in individuals with depression who also experience anxiety. This is particularly relevant given findings from the National Council of Social Service's 2025 Quality of Life study, which reported that 26 per cent of adults experience mild depressive or anxiety symptoms, 19 per cent moderate symptoms and 7 per cent severe symptoms.

The study, which aims to recruit 30 participants, will begin in October this year and run until early 2029. Moving beyond the conventional approach of targeting a single brain region commonly used for depression, the study will use two types of personalised stimulation targets, linked respectively to depression- and anxiety-related circuits.

Repeated fMRI and symptom assessments will allow researchers to investigate how different stimulation locations affect changes in the brain and in symptoms within individuals over time. Participants will have up to 48 hours of MRI scans over 10 months, providing an unusually detailed view of brain network changes within individuals over time. Insights from this work will support the development of more precise, individualised brain target maps to personalise future pTMS treatments in this patient group. The study is funded by the National Research Foundation, Singapore.

Assoc Prof Thomas Yeo said, "These studies represent an important step towards a more precise and individualised approach to mental health care. By combining brain imaging with computational modelling, we are able to better understand how each patient's symptoms are reflected in their brain circuitry, and translate that insight into targeted treatment. This integrated approach has the potential to reshape how we deliver TMS and improve outcomes for patients who have not benefitted from conventional methods."

Background

The work builds on an established NUH–NUS Medicine collaboration. The brain imaging analysis technology was developed at NUS and subsequently commercialised through B1Neuro, a local NUS spin-off. B1Neuro provides this technology to NUH and Mount Alvernia Hospital. Dr Ru Kong, Assoc Prof Thomas Yeo, and Adj Assoc Prof Tor Phern Chern serve as the Chief Scientific Officer, Scientific Advisor and Clinical Advisor respectively.