Breaking News
March 23, 2018 - Safe-sleep recommendations for infants have not reduced sudden deaths in newborns
March 23, 2018 - Survey finds inappropriate prescribing of antibiotics for hospitalized children
March 23, 2018 - Researchers propose alternative treatment to target lymphoma signaling at its root
March 22, 2018 - Compound found in beet extract could help slow progression of Alzheimer’s disease
March 22, 2018 - Lower temperatures can trigger the body’s ‘good’ fat formation at cellular level
March 22, 2018 - Sentinel lymph node biopsies could be safely avoided for some breast cancer patients
March 22, 2018 - Combined Preeclampsia Test Superior to U.K. Standards
March 22, 2018 - Exclusive breastfeeding in hospital associated with longer breastfeeding duration
March 22, 2018 - Researchers prove link between common childhood cancer and inflammation
March 22, 2018 - Targeting aberrantly active telomerase to treat therapy-resistant melanoma
March 22, 2018 - California’s tax on millionaires yields big benefits for people with mental illness, study finds
March 22, 2018 - Weight-loss surgery reduces risk for severe chronic kidney disease and kidney failure
March 22, 2018 - Study identifies two genes associated with extreme nausea and vomiting during pregnancy
March 22, 2018 - Biohaven Enrolls First Patient In Phase 3 Clinical Trial To Evaluate Rimegepant Zydis® ODT In The Acute Treatment Of Migraine
March 22, 2018 - ‘Bionic Pancreas’ OK for Very Young T1D Patients
March 22, 2018 - Parkinson’s gene initiates disease outside of the brain
March 22, 2018 - NMSU researchers publish findings on effective measures to remedy stress among adolescents
March 22, 2018 - Immunotherapy combination increases overall survival in people with kidney cancer, study shows
March 22, 2018 - Gut microbiome may promote pancreatic cancer by inducing immune suppression
March 22, 2018 - New ocular inserts allow patient’s cornea to absorb more antibiotics
March 22, 2018 - FDA Alert: NeuroBlate Probe by Monteris Medical: Letter to Health Care Providers, Class I Recall
March 22, 2018 - Cessation of exercise can result in increased depressive symptoms
March 22, 2018 - Morning Break: Booze Study Brouhaha; Stem Cells for MS; Big Debt Problem
March 22, 2018 - New wearable tech from Western may hold big benefits for people with Parkinson’s
March 22, 2018 - Immune cells can repopulate in the retina after elimination, mice study shows
March 22, 2018 - States extend Medicaid for birth control, cutting costs — and future enrollment
March 22, 2018 - Research provides better understanding of how cancerous cells behave in low oxygen
March 22, 2018 - Menopausal hormone therapy taken soon after menopause may benefit the brain
March 22, 2018 - Booze Boosts Your Heart Rate
March 22, 2018 - Skeptical Cardiologist: Classifying Heart Failure
March 22, 2018 - Instead of nagging your spouse to lose weight, try going on a diet yourself
March 22, 2018 - Neem Biotech to share findings on cystic fibrosis biofilm disruption at ECFS Basic Science Conference
March 22, 2018 - Study uncovers new genetic cause of posterior polymorphous corneal dystrophy
March 22, 2018 - ENDO: Big Breakfast May Help in Diabetics
March 22, 2018 - I’m not overweight, so why do I need to eat healthy foods?
March 22, 2018 - UCLA-led study suggests unexpected reason for reduction in cardiovascular health disparities
March 22, 2018 - Study suggests detailed neuropsychological assessment for brief cardiac arrest survivors
March 22, 2018 - Anticoagulant drugs found safe to use in patients undergoing surgery for irregular heartbeat
March 22, 2018 - SP Industries appoints Brian Larkin as new President and CEO
March 22, 2018 - GTx Announced New Data Demonstrating Enobosarm’s Potential to Treat Stress Urinary Incontinence
March 22, 2018 - Higher Risk of Brain Deficits in Older Alcoholics
March 22, 2018 - Top US health official resigns in conflict of interest
March 22, 2018 - Study shows benefits of hair loss drug in improving cognitive function and vascular health
March 22, 2018 - Researchers explain link between 2 key Alzheimer’s proteins
March 22, 2018 - Patients on replacement therapy with thyroid hormone may have more comorbidities
March 22, 2018 - Higher online patient ratings linked to urologists who saw fewer Medicare patients
March 22, 2018 - FDA Approves Ilumya (tildrakizumab-asmn) for the Treatment of Moderate-to-Severe Plaque Psoriasis
March 22, 2018 - Beer Raises Heart Rate; KardiaBand Hyperkalemia Test; CHD Clinics
March 22, 2018 - A retinal implant that is more effective against blindness
March 22, 2018 - New system based on artificial intelligence provides reliable detection of breast cancer
March 22, 2018 - Research offers new understanding about cause of Parkinson’s disease
March 22, 2018 - HORIBA’s Microsemi CRP analyzer improves quality of care in emergency pediatric units, study shows
March 22, 2018 - Neuroscientists move closer to developing tools for deciphering brain function
March 22, 2018 - New test methods with less fear
March 22, 2018 - Range of Vaginal Dryness Products Can Help Postmenopausal Women: Study
March 22, 2018 - Higher Dose Tx Deemed Safe in Pulmonary TB
March 22, 2018 - Discovery of new ALS gene points to cytoskeleton as potential target for drug development
March 22, 2018 - Diet soda associated with higher odds of diabetic retinopathy
March 22, 2018 - LSD reduces ‘sense of self’
March 22, 2018 - Vitamin D deficiency linked to metabolic syndrome in postmenopausal women
March 22, 2018 - Changes in the intestines may be responsible for reversal of diabetes after bariatric surgery
March 22, 2018 - iPads and Cancer; Clot Retrieval and Stroke: It’s PodMed Double T!
March 22, 2018 - Premature births linked to changes in mother’s bacteria
March 22, 2018 - Brain SPECT scans predict treatment outcomes in patients with depression
March 21, 2018 - Researchers succeed in integrating artificial organelles into cells of living organism
March 21, 2018 - Researchers discover ‘missing mutation’ in severe infant epilepsy
March 21, 2018 - Researchers develop statistics-based computational scheme to zoom in on brain function
March 21, 2018 - Verge joins Genomics England’s Discovery Forum industry partnership
March 21, 2018 - Trovagene Announces First Patient Successfully Completes Cycle 1 of Treatment with PCM-075 in Combination with Low Dose Cytarabine (LDAC) in AML Trial
March 21, 2018 - Congenital Cardiac Cath Tx Often Strays from Guidelines
March 21, 2018 - Marked increase in cardiovascular risk factors in women after preeclampsia
March 21, 2018 - New app may help predict, track manic and depressive episodes in bipolar patients
March 21, 2018 - Discovery of genes could lead to development of novel therapies for EBV-related cancers
March 21, 2018 - High-fat, high-cholesterol diet depletes ranks of artery-protecting immune cells
March 21, 2018 - Research misconduct allegations shadow likely CDC appointee
March 21, 2018 - Most Breast Ca Patients Fail to Get Genetic Counseling
March 21, 2018 - Lopsided ear function can lead to lopsided brain development
March 21, 2018 - Acupuncture helps manage menopausal symptoms, review finds
March 21, 2018 - Motor skill training may contribute to reading skills in obese children
March 21, 2018 - Poor dental health may be related to increased diabetes risk

Improving clinical trials with machine learning

image_pdfDownload PDFimage_print

 “Current statistical models are too simple. They fail to capture complex biological variations across people, discarding them as mere noise. We suspected this could partly explain why so many drug trials work in simple animals but fail in the complex brains of humans. If so, machine learning capable of modelling the human brain in its full complexity may uncover treatment effects that would otherwise be missed,” said the study’s lead author, Dr Parashkev Nachev (UCL Institute of Neurology).

To test the concept, the research team looked at large-scale data from patients with stroke, extracting the complex anatomical pattern of brain damage caused by the stroke in each patient, creating in the process the largest collection of anatomically registered images of stroke ever assembled. As an index of the impact of stroke, they used gaze direction, objectively measured from the eyes as seen on head CT scans upon hospital admission, and from MRI scans typically done 1-3 days later.

They then simulated a large-scale meta-analysis of a set of hypothetical drugs, to see if treatment effects of different magnitudes that would have been missed by conventional statistical analysis could be identified with machine learning. For example, given a drug treatment that shrinks a brain lesion by 70%, they tested for a significant effect using conventional (low-dimensional) statistical tests as well as by using high-dimensional machine learning methods.

The machine learning technique took into account the presence or absence of damage across the entire brain, treating the stroke as a complex “fingerprint”, described by a multitude of variables.

“Stroke trials tend to use relatively few, crude variables, such as the size of the lesion, ignoring whether the lesion is centred on a critical area or at the edge of it. Our algorithm learned the entire pattern of damage across the brain instead, employing thousands of variables at high anatomical resolution. We used well-established methods of machine learning, teaching the algorithm on subsets of data and then testing its performance on other subsets it had not seen,” explained the study’s first author, Tianbo Xu (UCL Institute of Neurology).

The advantage of the machine learning approach was particularly strong when looking at interventions that reduce the volume of the lesion itself. With conventional low-dimensional models, the intervention would need to shrink the lesion by 78.4% of its volume for the effect to be detected in a trial more often than not, while the high-dimensional model would more than likely detect an effect when the lesion was shrunk by only 55%.

“Conventional statistical models will miss an effect even if the drug typically reduces the size of the lesion by half, or more, simply because the complexity of the brain’s functional anatomy—when left unaccounted for—introduces so much individual variability in measured clinical outcomes. Yet saving 50% of the affected brain area is meaningful even if it doesn’t have a clear impact on behaviour. There’s no such thing as redundant brain,” said Dr Nachev.

The researchers say their findings demonstrate that machine learning could be invaluable to medical science, especially when the system under study—such as the brain—is highly complex.

“The real value of machine learning lies not so much in automating things we find easy to do naturally, but formalising very complex decisions. Machine learning can combine the intuitive flexibility of a clinician with the formality of the statistics that drive evidence-based medicine. Models that pull together 1000s of variables can still be rigorous and mathematically sound. We can now capture the complex relationship between anatomy and outcome with high precision,” said Dr Nachev.

“We hope that researchers and clinicians begin using our methods the next time they need to run a clinical trial,” said co-author Professor Geraint Rees (Dean, UCL Faculty of Life Sciences).

The study was funded by Wellcome and the National Institute for Health Research University College London Hospitals Biomedical Research Centre.

Tagged with:

About author

Related Articles