AI-Assisted Diagnosis and Management of Traumatic Brain Injury and Tinnitus : The Hearing Journal

Journal Logo


AI-Assisted Diagnosis and Management of Traumatic Brain Injury and Tinnitus

Hayes, Taylor

Author Information
The Hearing Journal 76(03):p 28,29, March 2023. | DOI: 10.1097/01.HJ.0000922296.59525.16
  • Free

For more than three decades, the Brain Injury Association of America (BIAA) has campaigned in observation of Brain Injury Awareness throughout March. BIAA works to educate on the prevention of traumatic brain injury (TBI), promote strategies that improve the quality of life for persons living with TBI and their families, and de-stigmatize brain injuries as it affects millions of people each year. TBIs are associated with many neurologic, psychological, and physical consequences that drastically impact an individual’s behavior, p hysical abilities, and overall quality of life. TBI is a significant global cause of mortality and morbidity with an increasing incidence, especially in low- and middle-income countries. 1 In the United States, 50,000 people die each year from TBI. 2 Depending on the severity of the TBI, patients may face another common health problem, tinnitus, that can last a few days or the rest of their lives.

F1 Artificial intelligence, traumatic brain injury, tinnitus.

While the term artificial intelligence (AI) was coined back in the 1960s, it’s now possible for machines to learn from experience, adjust to new inputs, and perform human-like tasks. 3 Today, AI is seen in familiar digital smart assistants like Siri, Alexa, and Google Assistant in addition to social media platforms and facial recognition features. AI is a rich realm of data, algorithms, analytics, deep learning, neural networks, and insights that’s constantly growing and adapting. 4 This great knowledge bank proves useful for the health care industry and its patients. With the latest technology, AI helps clinicians with medical decision-making, management, automation, administrative tasks, and workflows. 4 Due to the prevalence of TBI and mainstream media coverage of sports-related concussions/chronic traumatic encephalopathy, scientists have been making strides in the laboratory to improve AI diagnostic techniques and advance prognostic and recovery for TBI patients. Thus far, AI assistive technologies and aids have been developed and tested to improve brain rehabilitation therapy, make accurate diagnoses, reduce the cost of care, and increase the efficacy of therapy for patients.


TBI, also called craniocerebral trauma, can be caused by a bump, blow, jolt to the head, or penetrating injury (such as from a gunshot) to the head. 5 There are four main types and three levels of severity of TBI. A concussion is known as mild TBI and is the most common type of injury. Three out of four TBIs every year are concussions. 6 Brain contusions, penetrating brain injuries, and anoxic brain injuries are also types of TBI. Whether the injury is an internal bruise or results in a chemical imbalance, TBI may affect many parts of the body besides the head. Those with a TBI may experience physical, cognitive, and behavioral changes like nausea, dizziness, confusion, headaches, mood swings, or amnesia. Mild-to-moderate injuries usually require minimal treatment like rest and over-the-counter pain relievers to treat pain. 7 Since most patients who are admitted to the emergency department for severe TBIs typically undergo surgery as well, recovery from more severe TBIs require diligence. Treatment for TBI depends on severity but it might also include therapies like rehabilitation, cognitive behavioral therapy, anger management, and counseling, which are helpful for retraining the brain’s pathways and providing coping mechanisms.


Currently, physical examinations, imaging, blood tests, and assessments are used to diagnose TBI. However, traditional therapies, treatments, and diagnostic methods can be enhanced using AI and lead to better patient outcomes. For example, researchers at Helsinki University Hospital started to develop an AI-based algorithm that could help physicians treat patients with severe TBI. At its best, such an algorithm could predict the outcome of the individual patient and give objective data regarding the condition and prognosis of the patient and how it changes during treatment. 8 Another study at Imperial College London and Cambridge developed an AI algorithm that can detect and identify different types of brain injuries and brain lesions using images of CT scans. 9 While researchers at Rice University aim to create a data-driven framework using wearable devices that measure bioelectrical signal recordings for systematic detection of mild TBI or concussions. The data collected from this framework and new processing routines could prove useful in future smart helmet technologies and may serve as a first step toward uncovering techniques for signal modulation-based treatment of concussions. 10 All of these research projects are helping to improve TBI diagnosis techniques using algorithms. AI has also been tested for patient monitoring and predicting surgery outcomes.


A systematic literature review that evaluated studies on PubMed from July and August 2012 revealed up to 53% of individuals suffering from TBIs develop tinnitus. 11 Kreuzer et al. show the delay between trauma and tinnitus onset is in the range of several days in most cases but may last up to 4 to 12 weeks. 11 When symptoms last more than six months, it’s known as chronic tinnitus which is manageable but incurable. While the type of sound varies per individual, tinnitus is categorized as the constant sound of ringing or buzzing in the ears when no external sound is present. Tinnitus occurs when hair cells, which help transform sound waves into electrical signals that travel through the auditory nerve, are damaged by loud noise or medication. In turn, the brain doesn’t receive the signals it’s expecting which stimulates abnormal activity in the neurons and creates the illusion of sound. Tinnitus can also be caused by acute excessive noise exposure, whiplash, neck trauma, blast injury, medications that damage the nerves in the ear, impacted earwax, middle ear infections, and aging. 12 More than 200 drugs are known to cause tinnitus when you start or stop taking them. 12

Specific treatment recommendations for continual tinnitus varies per person but may include hearing aids (for those with hearing loss), sound therapy, or behavioral therapy. A general assessment of overall health including diet, exercise, and mental health check-ins are essential observations as the intensity of tinnitus can fluctuate depending on these factors. 13 Avoiding triggers such as loud noises or discontinuing certain medication can reduce the effects or prevent it from getting worse. While time-consuming, a common combination of sound-based therapy and counseling therapy is usually provided by audiologists. 13


In the updated Health app on iOS devices, a noise level measurement feature measures noise levels in the environment and headphones to determine if levels are safe. If a person is exposed to loud environments or listen to music at a high volume, the app will send a warning. Additionally, as of April 2021, a search on the Apple iTunes store identified over 100 digital albums identified as “tinnitus relief,” which provide noises that balance the “ear ringing” sounds. 14 While app technology can help with treatment and lifestyle changes for those dealing with tinnitus, a new generation boasts wearable technology with physiological sensors, multiple therapy options, AI-personalized therapies, and new diagnostic techniques. Since tinnitus is usually diagnosed based on patient-reported symptoms and experiences alone, Mehrnaz Shoushtarian at the Bionics Institute and colleagues in Melbourne, Australia, have developed an algorithm based on the results of brain imaging that can detect whether a person has tinnitus, and also how severe it is. 15 The AI can spot the presence of tinnitus with 78% accuracy and distinguish between mild and severe forms with 87% accuracy. 15 Similarly, a 2019 study performed by Liu et al. of 46 tinnitus patients at Beijing Friendship Hospital revealed an accuracy of 80% in distinguishing between tinnitus patients and healthy subjects using neuroimaging biomarkers from machine learning features. 16 Their results show that thirteen brain regions can effectively be used to differentiate patients with tinnitus from healthy subjects. 16 The high accuracy levels of both these studies provide solid groundwork for the future of tinnitus diagnosis. For those already diagnosed with tinnitus, a pilot study led by James A. Henry developed Progressive Tinnitus Management (PTM), which uses education and counseling to help patients learn how to self-manage their reactions to tinnitus. 17 Participants received telephone counseling by an audiologist and a psychologist. Participants were surveyed on their improvement in self-perceived functional limitations due to tinnitus. Not only has AI detected internal characteristics of tinnitus, but it has also improved quality of life for patients.


As AI evolves, it magnifies the ability to use personal information in ways that can intrude on privacy interests by raising analysis of personal information to new levels of power and speed. 18 In health care, AI can be used to make care decisions using all the available information, but some have concerns about patient privacy. Due to the unclear laws and data breaches associated with certain aspects of AI and large data companies, AI tools are less often utilized in real-world medical settings due to a lack of transparency and trustworthiness. This also contributes to the limited agreement on the most useful algorithms and types of bioelectrical features for the prediction and diagnosis of TBI, stunting AI growth and research in this field.


Since TBI is still not well understood in the research laboratory, nor the clinical setting, it will take time until prognostic models and algorithms can be widely implemented into daily clinical practice but there is hope for the future. Despite these concerns, it’s important to consider that AI tools can minimize patient morbidity and decrease costs once widely implemented. While the data still must be evaluated by medical professionals, using AI in medical diagnosis has shown immense promise in changing the standards of medical care for those affected by complex and uncertain TBI and tinnitus cases.


1. University of Helsinki "Artificial intelligence-based algorithm for intensive care of traumatic brain injury" ScienceDaily. ScienceDaily, November 2019. Retrieved from:
2. Division of Acute Care, Rehabilitation Research, and Disability Prevention, National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, U.S. Department of Health and Human Services “Report to Congress: Traumatic Brain Injury in the United States” CDC, December 1999. Retrieved from:,50%2C000%20people%20die
3. SAS® Insights “Artificial Intelligence: What it is and why it matters”. Retrieved from:
4. Long, Marlee “Artificial Intelligence in Medical Diagnosis” Aidoc. November 2020. Retrieved from
5. CDC “Get the Facts About TBI” Retrieved from
6. Cleveland Clinic “Traumatic Brain Injury” 2021. Retrieved from:
7. Mayo Clinic “Traumatic brain injury”. Retrieved from
8. Benjamin Y, Gravesteijn DN, Ercole A, et al. 2020 Machine learning algorithms performed no better than regression models for prognostication in traumatic brain injury Journal of Clinical Epidemiology
9. Monteiro M, Newcombe V, Mathieu F, et al. 2020 Multiclass semantic segmentation and quantification of traumatic brain injury lesions on head CT using deep learning: an algorithm development and multicentre validation study The Lancet
10. Aazhang Behnaam Wearable Technologies and Machine Learning Methodologies for Systematic Detection of Mild Traumatic Brain Injuries. Retrieved from:
11. Kreuzer PM, Landgrebe M, Vielsmeier V, Kleinjung T, De Ridder D, Langguth B Trauma-associated tinnitus. Journal of Head Trauma Rehabilitation. September/October 2014.
12. National Institute on Deafness and Other Communication Disorders “Tinnitus” February 2014. Retrieved from
13. American Tinnitus Association “Managing your Tinnitus”. Retrieved from:
14. Grant D, Sanders PJ, Doborjeh Z, Doborjeh M, Boldu R, Sun K, Barde A 2021 A state-of-art review of digital technologies for the next generation of tinnitus therapeutics Frontiers in Digital Health
15. Shoushtarian M, Alizadehsani R, Khosravi A, et al. 2020 Objective measurement of tinnitus using functional near-infrared spectroscopy and machine learning PLoS ONE
16. Yawen L, Haijun N, Jianming Z 2019 Morphological Neuroimaging Biomarkers for Tinnitus: Evidence Obtained by Applying Machine Learning Neural Plasticity
17. Henry J, Schechter M, Zaugg T, Myers P Progressive audiologic tinnitus management. ASHA Wire, 2008
18. Kerry Cameron “Protecting privacy in an AI-driven world” Brookings. Retrieved from:
Copyright © 2023 Wolters Kluwer Health, Inc. All rights reserved.