Can AI be utilized to identify brain damage following a stroke?

 Once you identify the specific area where the damage is; WHAT IS THE EXACT REHAB PROTOCOL TO BE USED TO RECOVER FROM THAT DAMAGE? Totally incomplete research!

Can AI be utilized to identify brain damage following a stroke?

 

The study examined the potential of artificial intelligence (AI), specifically the large language model GPT-4, to help locate brain lesions resulting from strokes. These lesions play a crucial role in predicting the long-term effects of a stroke and determining the appropriate treatment and prognosis for affected individuals. The research, conducted by Dr. Jung-Hyun Lee from SUNY Downstate Health Sciences University, utilized text from health histories and neurologic exams of 46 stroke patients to train GPT-4 in accurately identifying the location and extent of lesions in the brain.

The findings of the study revealed that GPT-4 was successful in locating lesions in the brains of many participants, determining the side of the brain affected as well as the specific brain region, with the exception of lesions in the cerebellum and spinal cord. The AI model demonstrated a sensitivity of 74% and a specificity of 87% in identifying the side of the brain with lesions, and a sensitivity of 85% and a specificity of 94% in pinpointing the brain region involved. Additionally, GPT-4 showed consistency in its results for the number of brain lesions, side of the brain, and brain regions in a majority of cases.

Although GPT-4 was able to provide accurate answers for 41% of participants when combining responses to all three questions across all three times, the study notes that further refinement and validation are needed before its clinical use. A key limitation of the study is that the accuracy of GPT-4 relies on the quality of information it receives, and detailed health histories and neurologic exam information may not always be available for all stroke patients. However, the potential of AI models like GPT-4 to assist in locating brain lesions after a stroke is seen as promising, particularly in underserved regions where access to neurologic care is limited.

The study highlights the importance of accurate identification of brain lesions following a stroke, as this information can significantly impact the long-term outcomes and treatment strategies for affected individuals. By leveraging the capabilities of AI models like GPT-4 to analyze health histories and neurologic exam data, neurologists may be able to streamline the diagnostic process and improve the efficiency of lesion localization. This advancement has the potential to reduce disparities in healthcare access and delivery, especially in regions where neurologic care is scarce.

Moving forward, further research and validation are needed to enhance the accuracy and reliability of AI models like GPT-4 in locating brain lesions after a stroke. As technology continues to evolve, the integration of AI in neurology practice holds promise for improving patient outcomes and enhancing the accessibility and quality of healthcare services worldwide. Continued collaboration between neurologists, researchers, and AI experts will be crucial in harnessing the full potential of artificial intelligence in advancing neurologic care and addressing the global health challenges posed by strokes and other neurological conditions.

Reproductive History Linked to Later Stroke Risk

This is useless information until there are protocols that prevent this stroke risk. Totally incomplete research. 

Reproductive History Linked to Later Stroke Risk

Infertility, stillbirth, miscarriage may increase stroke risk, analysis of eight studies suggests

by Elizabeth Short, Editorial Intern, MedPage Today June 22, 2022

A history of infertility, recurrent miscarriage, or stillbirth may be a risk factor for stroke later in life, according to an analysis of eight prospective cohort studies.

Among over 600,000 women, infertility was associated with an increased risk of non-fatal stroke (HR 1.14, 95% CI 1.08-1.20), while a history of at least three miscarriages was associated with higher risks of both non-fatal stroke (HR 1.35, 95% CI 1.27-1.44) and fatal stroke (HR 1.82, 95% CI 1.58-2.10), reported Gita Mishra, PhD, of the University of Queensland in Australia, and colleagues.

Furthermore, those who experienced a stillbirth were at a 31% higher risk of non-fatal stroke, and those who had a history of recurrent stillbirth were at a 26% higher risk of fatal stroke, they noted in The BMJ.

"A history of recurrent miscarriages and death or loss of a baby before or during birth could be considered a female specific risk factor for stroke, with differences in risk according to stroke subtypes," Mishra and team concluded. "These findings could contribute to improved monitoring and stroke prevention for women with such a history."

Analyses by subtypes of non-fatal stroke showed infertility was associated with an increased risk of ischemic stroke (HR 1.15, 95% CI 1.07-1.23), while women with recurrent miscarriage were more likely to experience ischemic and hemorrhagic stroke versus women without miscarriage (HR 1.37, 95% CI 1.23-1.53, and HR 1.41, 95% CI 1.08-1.84, respectively).

As for fatal stroke, women with recurrent miscarriages were more likely to experience ischemic and hemorrhagic fatal stroke (HR 1.83, 95% CI 1.39-2.41, and HR 1.84, 95% CI 1.39-2.44, respectively), and those with recurrent stillbirth were more likely to have hemorrhagic fatal stroke (HR 1.44, 95% CI 1.35-1.53).

Mishra and team noted that the link between infertility and increased stroke risk may be due to disorders such as polycystic ovary syndrome and premature ovarian insufficiency, while endothelial dysfunction may explain the increased risk of stroke for women with a history of recurrent stillbirth or miscarriage.

For this analysis, Mishra and colleagues analyzed data on 618,851 women ages 32 to 73 from eight studies from seven countries -- China, Sweden, the Netherlands, the U.K., Japan, Australia, and the U.S. -- as part of the InterLACE (International Collaboration for a Life Course Approach to Reproductive Health and Chronic Disease Events) consortium. Of the included women, 275,863 had data on fatal and non-fatal stroke: 9,265 (2.8%) experienced a non-fatal stroke and 4,003 (0.7%) experienced a fatal stroke.

Median follow-up was 13 years after a non-fatal stroke and 9.4 years after a fatal stroke. Average ages at the time of first non-fatal stroke and fatal stroke were 62 and 71.

Of the included women, 17.2% experienced infertility, 16.6% experienced miscarriage, and 4.6% experienced stillbirth.

There were several limitations to this analysis, the authors acknowledged. Since data on infertility, miscarriage, and stillbirth were collected from questionnaires, recall bias is possible. Furthermore, while most models adjusted for certain comorbidities, others, such as endometriosis, thyroid disorders, and pelvic inflammatory disease, were not available in all studies.

• 

  Elizabeth Short is an editorial intern for MedPage Today. She is based in New York City. Follow

Disclosures

This study was funded by the Australian National Health and Medical Research Council Centres of Research Excellence.

The study authors reported no conflicts of interest.

Primary Source

The BMJ

Source Reference: Liang C, et al "Infertility, recurrent pregnancy loss, and risk of stroke: pooled analysis of individual patient data of 618851 women" BMJ 2022; DOI: 10.1136/bmj-2022-070603.

Synopsi

Journal of Clinical Oncology

Burden of Long-Term Morbidity Borne by Survivors of Acute Myeloid Leukemia Treated With Blood or Marrow Transplantation: The Results of the BMT Survivor Study.

AIDS

Associations between HIV viremia during antiretroviral therapy and cardiovascular disease.

Circulation

Cost-Effectiveness of Catheter Ablation Versus Antiarrhythmic Drug Therapy in Atrial Fibrillation: The CABANA Randomized Clinical Trial.

 

Shoulder and elbow muscle activity during fully supported trajectory tracking in people who have had a stroke

 You researched something. How EXACTLY will this be used to get survivors recovered? Incomplete research. I would have the mentors and senior researchers fired for not specifying research goals correctly.  A lot of dead wood in stroke needs to be removed.

Shoulder and elbow muscle activity during fully supported trajectory tracking in people who have had a stroke

 A.M. Hughes a,*
, C.T. Freeman b, 

J.H. Burridge a, 

P.H. Chappell b, 

P.L. Lewin b, 

E. Rogers b
a School of Health Sciences, University of Southampton, Southampton, SO17 1BJ, UK
b School of Electronics and Computer Science, University of Southampton, Southampton, SO17 1BJ, UK

 abstract

An inability to perform tasks involving reaching is a common problem for stroke patients. This paper provides an insight into mechanisms associated with recovery of upper limb function by examining how stroke participants’ upper limb muscle activation patterns differ from those of neurologically intact participants, and how they change in response to an intervention.In this study, five chronic stroke participants undertook nine tracking tasks in which trajectory (orientation and length), speed and resistance to movement were varied. During these tasks, EMG signals were recorded from triceps, biceps, anterior deltoid, upper, middle and lower trapezius and pectoralis major.Data collection was performed in sessions both before, and after, an intervention in which participants performed a similar range of tracking tasks with the addition of responsive electrical stimulation applied to their triceps muscle. The intervention consisted of eighteen one hour treatment sessions, with two participants attending an additional seven sessions. During all sessions, each participant’s arm was supported by a hinged arm-holder which constrained their hand to move in a two dimensional plane.Analysis of the pre intervention EMG data showed that timing and amplitude of peak EMG activity for all stroke participants differed from neurologically intact participants. Analysis of post intervention EMG data revealed that statistically significant changes in the sequantities had occurred towards those of neurologically intact participants.