DR. JAMES CUMMINGS DUNSTAN JR., M.D.

LYNCHBURG, VA

Research Active
Physical Medicine & Rehabilitation NPI registered 21+ years 50 publications 2016 – 2026 NPI: 1750384137
PrognosisNeoplasmsMagnetic Resonance ImagingBrainMelanomaSkin NeoplasmsBreast NeoplasmsSex FactorsChileLymphocytes, Tumor-InfiltratingNatural Language ProcessingPeruWhite MatterReferral and ConsultationDyslexia

Practice Location

3300 RIVERMONT AVE
LYNCHBURG, VA 24503-2030

Phone: (434) 947-4651

What does JAMES DUNSTAN research?

Dr. Dunstan studies a range of topics, but particularly emphasizes the mental health of young people and brain development in early childhood. He has researched how chatbot technology can significantly aid young individuals dealing with emotional issues, showing promise as an accessible resource for support. Additionally, he investigates the use of advanced imaging techniques like MRI in understanding brain functions in children aged 3 to 5 years. His work also extends to analyzing workplace accidents and improving cancer diagnosis through automated systems, aiming to enhance safety and treatment outcomes in various populations.

Key findings

  • Using new MRI techniques, Dr. Dunstan improved the success rates of brain scans in children aged 3 to 5 years, leading to better data for understanding brain development.
  • Participants using the chatbot MYLO for mental health support reported a 44% reduction in distress levels (Cohen d=-1.07) after four weeks.
  • In analyzing 350,000 health reports on workplace accidents, it was found that 57.3% of the accidents involved men, highlighting significant gender differences in occupational safety.
  • An automated system decoding tumor descriptions in pathology reports achieved up to 90% accuracy, streamlining cancer diagnosis in Spanish-speaking populations.
  • Research on the hippocampus revealed boys with depressed mothers had smaller brain volumes, indicating early brain features linked to depression risk in children.

Frequently asked questions

Does Dr. Dunstan study childhood brain development?
Yes, he examines brain structure and function in young children using advanced MRI techniques.
What mental health treatments has Dr. Dunstan researched?
He has studied a mental health chatbot designed to support young people, showing significant improvements in their distress and anxiety.
Is Dr. Dunstan's work relevant to workplace safety?
Yes, he analyzes workplace accident data to identify gender-specific risks and improve safety measures for employees.
How does Dr. Dunstan's research help children at risk for depression?
His studies help identify brain and social risk factors in children that may indicate a higher likelihood of developing depression.
What advancements has Dr. Dunstan made in cancer treatment?
He developed an automated system for coding tumors in pathology reports, contributing to more efficient cancer diagnosis and care.

Publications in plain English

Scanning Dos and Don'ts: Using Magnetic Resonance Imaging in Awake Children Aged 3 to 5 Years to Assess Brain Structure and Function.

2026

Journal of visualized experiments : JoVE

Dunstan J, Ratliff EL, Riggins T

Plain English
This study looked at how to successfully use MRI scans to examine the brain structure and function of young children aged 3 to 5 years. Researchers found that by using new techniques, they could significantly improve the accuracy of scans, leading to better quality data from these young participants. For example, the protocol they developed increased the success rate of scans, helping to better understand how children’s brains develop during important early years. Who this helps: This study benefits researchers, doctors, and ultimately young patients by providing clearer insights into brain development.

PubMed

Developing and Validating an Automatic Support System for Tumor Coding in Pathology Reports in Spanish.

2025

JCO clinical cancer informatics

Villena F, Báez P, Peñafiel S, Rojas M, Paredes I +1 more

Plain English
This study developed a new automatic system to help code descriptions of tumors in pathology reports written in Spanish, which are often hard to interpret manually. The system was tested on 1,101 documents and achieved high accuracy, especially in recognizing details about tumor types and locations, with scores of 0.86 and 0.90 for different aspects. This matters because it can save time for healthcare professionals and improve the accuracy of cancer data collection, which helps in better cancer treatment and research. Who this helps: This benefits doctors, cancer registries, and researchers working to improve cancer care.

PubMed

Hippocampal subregion volumes and preadolescent depression risk in the ABCD sample.

2025

Journal of affective disorders

Parker AJ, Sorcher LK, Cutshaw OP, Botdorf M, Dunstan J +2 more

Plain English
This study looked at how different parts of the hippocampus, a brain area important for depression, relate to the risk of depression in kids aged 9 to 10. It found that boys with a mother who had depression had smaller volumes in a part of the hippocampus, and that kids from lower socioeconomic backgrounds with larger volumes in another hippocampal area showed more internalizing symptoms (like anxiety and sadness) both initially and two years later. Understanding these links is crucial because it helps identify which children might be at greater risk for developing depression. Who this helps: This helps parents and mental health professionals identify at-risk children early on.

PubMed

NLP modeling recommendations for restricted data availability in clinical settings.

2025

BMC medical informatics and decision making

Villena F, Bravo-Marquez F, Dunstan J

Plain English
This study looked at how well different artificial intelligence models can analyze unstructured text data in healthcare when there is limited data available. The researchers found that specialized language models for clinical data worked best; they achieved an impressive 88.85% accuracy for prioritizing patient referrals and 53.79% for classifying the type of referral needed. These results matter because they can help improve decision-making in healthcare, ensuring patients get the care they need more efficiently. Who this helps: This helps doctors and healthcare providers.

PubMed

Sex differences in work-related accidents extracted from free text in Spanish using natural language processing.

2025

BMC public health

Dunstan J, Campaña-Herrera V, Miranda L, Ladrón de Guevara R, Pincheira P +2 more

Plain English
This study examined work-related accidents in Chile using advanced computer techniques to analyze 350,000 health reports. It found that 57.3% of the accidents involved men, while 42.7% involved women, with men mostly working as operators and women in cleaning roles. Men frequently suffered accidents involving sharp objects, while women often fell from the same height. This research is important because it highlights gender differences in workplace accidents and can help improve safety measures. Who this helps: This benefits workers, employers, and safety regulators in understanding and addressing workplace risks.

PubMed

A Rule-Based Conversational Agent for Mental Health and Well-Being in Young People: Formative Case Series During the Rise of Generative AI.

2025

JMIR formative research

Wrightson-Hester AR, Anderson G, Dunstan J, McEvoy PM, Sutton CJ +8 more

Plain English
A rule-based mental health chatbot called MYLO, co-designed with young people, was tested over four weeks with 24 young Australians with anxiety or depression, partly to see how it compared with users' experiences of newer AI tools like ChatGPT. Participants showed moderate to large improvements in distress, anxiety, and functioning despite many expecting more human-like conversation from the chatbot. The findings highlight the need to either upgrade rule-based chatbots or clearly communicate their limitations to users who now have high expectations from generative AI.

PubMed

Sequence Processing in Music Predicts Reading Skills in Young Readers: A Longitudinal Study.

2024

Journal of learning disabilities

Andrade PE, Müllensiefen D, Andrade OVCA, Dunstan J, Zuk J +1 more

Plain English
This study looked at how music skills, particularly in rhythm and timing, relate to reading abilities in young children. The researchers tested 45 Portuguese-speaking children, finding that their performance on a music transcription task (which focused on sequence processing) was just as effective as traditional phonological awareness measures in predicting their reading success over several years. Specifically, the music task accurately identified struggling readers as well as established methods used to diagnose reading disabilities. Who this helps: This research benefits teachers and parents by providing a new tool to identify children who might struggle with reading.

PubMed

Automatic Detection of Distant Metastasis Mentions in Radiology Reports in Spanish.

2024

JCO clinical cancer informatics

Ahumada R, Dunstan J, Rojas M, Peñafiel S, Paredes I +1 more

Plain English
This study focused on finding mentions of distant cancer spread (metastasis) in radiology reports written in Spanish. Researchers developed a system that can automatically identify if cancer is present in reports for prostate, colorectal, and breast cancers, achieving a high accuracy score of 0.92 for reports without metastasis and 0.90 for reports with it. This is important because it helps doctors quickly understand patient conditions, which can improve treatment planning and patient care. Who this helps: Patients with cancer and their doctors benefit from more accurate and efficient information about metastasis.

PubMed

The effect of age on musculoskeletal injury compensation costs in South Australian professional urban firefighters: A cohort study.

2024

Work (Reading, Mass.)

Dunstan J, Stenner B, Carrick-Ranson G, Norton K

Plain English
This study looked at how age affects the costs of injuries among professional urban firefighters in South Australia. It found that older firefighters, particularly those aged 60-70, had claims that cost over ten times more than those from younger firefighters aged 20-29, and they also lost more workdays when injured. This matters because as the workforce ages, fire services may face much higher expenses related to worker injuries, making it crucial for them to create better management and prevention strategies. Who this helps: This helps fire service organizations and their employees.

PubMed

Physics-informed neural networks for parameter estimation in blood flow models.

2024

Computers in biology and medicine

Garay J, Dunstan J, Uribe S, Sahli Costabal F

Plain English
This study looked at a new way to estimate blood flow parameters in the aorta using advanced computer algorithms called physics-informed neural networks (PINNs). The researchers tested these algorithms with noisy data from two types of blood flow: steady and changing. They found that PINNs provided reliable estimates, especially when there were many parameters to track, outperforming traditional methods. Who this helps: This benefits doctors and researchers working on cardiovascular health by improving how they understand blood flow dynamics.

PubMed

Unveiling Melanoma: A Deep Dive into Disparities at a Latin-American Cancer Institute.

2024

Annals of surgical oncology

Ziegler-Rodriguez G, De La Cruz-Ku G, Piedra-Delgado L, Torres-Maldonado J, Dunstan J +6 more

Plain English
This research focused on melanoma patients at a cancer center in Peru over 20 years. The study found that more patients were diagnosed with late-stage melanoma, with a significant increase in stage IV cases from 12.3% in 2005-2006 to 28.4% in 2017-2018. Additionally, survival rates for patients in stages II and III decreased significantly, indicating that fewer patients are surviving as time goes on, especially since half of the population lacks access to cancer care. Who this helps: This information is crucial for healthcare providers and policymakers to improve cancer treatment and access in underserved areas.

PubMed

ASO Author Reflections: Distilling Wisdom From Two Decades of Cutaneous Malignant Melanoma at a Peruvian Cancer Institute: A Stirring Call for Action.

2024

Annals of surgical oncology

Ziegler-Rodriguez G, Ziegler-Rodriguez O, De La Cruz-Ku G, Cotrina-Concha JM, Dunstan J +3 more

PubMed

A pseudonymized corpus of occupational health narratives for clinical entity recognition in Spanish.

2024

BMC medical informatics and decision making

Dunstan J, Vakili T, Miranda L, Villena F, Aracena C +4 more

Plain English
This study created a set of over 1,700 anonymized accounts of work-related accidents and illnesses in Spanish to help improve systems that recognize important medical terms. They tested an existing model that identifies diseases, body parts, and medications, finding that it performed well on these anonymized texts. This is important because it helps protect patient privacy while still allowing for the analysis of health data in a meaningful way. Who this helps: This benefits researchers and healthcare professionals working with patient data.

PubMed

Correction: Unveiling Melanoma: A Deep Dive into Disparities at a Latin-American Cancer Institute.

2024

Annals of surgical oncology

Ziegler-Rodriguez G, De La Cruz-Ku G, Piedra-Delgado L, Torres-Maldonado J, Dunstan J +6 more

PubMed

Response to Kempf et al on Methodological and Practical Aspects of a Distant Metastasis Detection Model.

2024

JCO clinical cancer informatics

Ahumada R, Dunstan J, Paredes I, Báez P

PubMed

Quantized current steps due to the synchronization of microwaves with Bloch oscillations in small Josephson junctions.

2024

Nature communications

Shaikhaidarov RS, Kim KH, Dunstan J, Antonov I, Golubev D +2 more

Plain English
This research studied how small superconducting devices called Josephson junctions behave when exposed to microwaves, specifically looking at a phenomenon called current quantization. They discovered that these junctions can show a specific kind of current behavior, known as Shapiro steps, when microwaves are synchronized with the junctions, achieving this at microwave frequencies up to 24 GHz and measuring a current of 7.7 nanoamperes. This finding is significant because it can help establish a new standard for measuring electric current, which is crucial for advancements in technology and scientific research. Who this helps: This benefits scientists and engineers working in quantum technology and precision measurement.

PubMed

Influence of sleep and nap habituality on mnemonic discrimination in early childhood: An online study.

2024

Cognitive development

Dunstan J, Canada KL, Spencer RMC, Riggins T

Plain English
This study looked at how naps and overnight sleep affect the ability of young children (ages 3 to 5) to remember and distinguish between similar items. Researchers found that children who regularly take naps showed poorer detail-recognition after napping but improved their memory after sleeping overnight. In contrast, children who don’t typically nap did not show any changes in memory after either napping or staying awake. This matters because it highlights the importance of overnight sleep for children who nap regularly to help them remember details, while those who don't nap may have enough brain development to remember accurately without additional sleep. Who this helps: This helps children who take regular naps and their parents.

PubMed

Correction: An Artificial Therapist (Manage Your Life Online) to Support the Mental Health of Youth: Co-Design and Case Series.

2024

JMIR human factors

Wrightson-Hester AR, Anderson G, Dunstan J, McEvoy PM, Sutton CJ +7 more

PubMed

Predicting no-show appointments in a pediatric hospital in Chile using machine learning.

2023

Health care management science

Dunstan J, Villena F, Hoyos JP, Riquelme V, Royer M +2 more

Plain English
This study looked at why so many patients miss their appointments at a pediatric hospital in Chile, where the no-show rate is around 20%, and sometimes as high as 29% for certain specialties like ophthalmology. Researchers used advanced computer methods to identify patterns that predict these no-shows based on factors like patients’ backgrounds and past attendance. They found that a targeted reminder system led to a significant drop in missed appointments by 10.3 percentage points, showing that understanding a patient's history and circumstances can help improve attendance. Who this helps: This benefits hospitals and doctors by reducing missed appointments and improving healthcare efficiency.

PubMed

An Artificial Therapist (Manage Your Life Online) to Support the Mental Health of Youth: Co-Design and Case Series.

2023

JMIR human factors

Wrightson-Hester AR, Anderson G, Dunstan J, McEvoy PM, Sutton CJ +7 more

Plain English
This study looked at an artificial intelligence chatbot called Manage Your Life Online (MYLO), designed to help young people aged 16 to 24 with mental health issues. Over two weeks, 13 participants found MYLO easy to use and beneficial, reporting a significant decrease in distress related to their problems and an improvement in their abilities to resolve conflicts. The feedback was overwhelmingly positive, indicating that MYLO could be a valuable tool for supporting youth mental health during times when traditional services are limited. Who this helps: Young people struggling with mental health issues.

PubMed

Distinctive neural correlates of phonological and reading impairment in fetal alcohol-exposed adolescents with and without facial dysmorphology.

2022

Neuropsychologia

Yu X, Dunstan J, Jacobson SW, Molteno CD, Lindinger NM +4 more

Plain English
This study looked at how brain activity and structure relate to reading difficulties in adolescents who were exposed to alcohol before birth. Researchers scanned the brains of 26 adolescents with facial differences from fetal alcohol syndrome, 29 with heavy alcohol exposure but no facial differences, and compared them to 19 typically developing adolescents. They found that those with facial characteristics showed more brain activity in a certain area during reading tasks, indicating they process reading differently, while those without facial differences had some better brain organization for reading but less efficient brain function overall. Who this helps: This helps educators and healthcare providers working with adolescents affected by fetal alcohol exposure.

PubMed

Human Papillomavirus, Cytomegalovirus Infection and P16 Staining in Breast Tumors from Peruvian Women.

2022

Asian Pacific journal of cancer prevention : APJCP

Calderon G, Castaneda CA, Castillo M, Sanchez J, Bernabe L +11 more

Plain English
This study examined breast cancer samples from Peruvian women to see how often certain viral infections, like human papillomavirus (HPV) and cytomegalovirus (HCMV), occur and how they relate to cancer features. They found HCMV in 72.5% of the tumors, while HPV was present in just 2.9%. The presence of the protein p16 was found in 28.15% of the cases and was linked to more aggressive cancer characteristics. Who this helps: This research benefits doctors and researchers working with breast cancer patients in Peru.

PubMed

Socioeconomic disadvantage and episodic memory ability in the ABCD sample: Contributions of hippocampal subregion and subfield volumes.

2022

Developmental cognitive neuroscience

Botdorf M, Dunstan J, Sorcher L, Dougherty LR, Riggins T

Plain English
The study looked at how economic disadvantage affects memory in children aged 9 to 13 by examining the size of specific areas in the brain known to relate to memory. Researchers found that higher levels of socioeconomic disadvantage were linked to smaller volumes in important brain areas and poorer memory performance over two years, with 66% of the children showing a decline in memory improvement. This is important because it highlights how economic factors can impact brain development and cognitive abilities in children. Who this helps: This helps children from low-income families by emphasizing the need for supports that could improve their cognitive health.

PubMed

Home language and literacy environment and its relationship to socioeconomic status and white matter structure in infancy.

2022

Brain structure & function

Turesky TK, Sanfilippo J, Zuk J, Ahtam B, Gagoski B +7 more

Plain English
This study looked at how the language and reading environment at home for infants affects brain development, specifically in areas that are important for reading skills. Researchers found that a richer home language and literacy environment is linked to better white matter organization in the brain, particularly in pathways related to reading. For example, infants with a more stimulating language environment showed greater brain development in certain areas related to reading. Understanding these connections early on can help develop better support strategies for infants to promote healthy brain growth. Who this helps: This helps infants and their families by promoting early literacy and language development.

PubMed

Training and intrinsic evaluation of lightweight word embeddings for the clinical domain in Spanish.

2022

Frontiers in artificial intelligence

Chiu C, Villena F, Martin K, Núñez F, Besa C +1 more

Plain English
This research focused on creating and evaluating lightweight word representations specifically for the Spanish language in the healthcare field. The team built a specialized database from various medical sources and used it to develop word embeddings, which are numerical representations that help computers understand medical text. They confirmed the quality of these embeddings through bilingual clinician evaluations and found that their simpler static versions can be effectively used in clinical settings without needing advanced technology. Who this helps: This benefits healthcare professionals who work with Spanish-speaking patients and need better tools for processing medical information.

PubMed

Patterns of Neural Functional Connectivity in Infants at Familial Risk of Developmental Dyslexia.

2022

JAMA network open

Yu X, Ferradal S, Dunstan J, Carruthers C, Sanfilippo J +6 more

Plain English
This study examined how the brain networks of infants who have a family history of dyslexia differ from those without such a history. Researchers found that 35 infants with a family history of dyslexia displayed unusual brain connectivity patterns compared to 63 infants without this history, particularly in a brain region linked to recognizing words. Understanding these differences early on is crucial because it might lead to better identification and support for children at risk of reading difficulties. Who this helps: This helps infants at risk for dyslexia and the parents and educators who support them.

PubMed

Association between PIK3CA Mutations in Blood and Tumor-Infiltrating Lymphocytes in Peruvian Breast Cancer Patients.

2022

Asian Pacific journal of cancer prevention : APJCP

Castaneda CA, Castillo M, Bernabe LA, Suarez N, Romero A +16 more

Plain English
This study looked at the DNA from blood samples of 183 breast cancer patients in Peru to see how it relates to features of their tumors, particularly the number of immune cells called tumor-infiltrating lymphocytes (TILs). They found that 35% of patients had a specific mutation in their DNA (called PIK3CA), which was linked to lower levels of immune cells and tended to indicate worse outcomes, including higher chances of cancer recurrence and shorter survival, although the results were not definitive. Understanding this link helps doctors tailor treatments and monitor patients more effectively. Who this helps: This helps patients with breast cancer and their doctors.

PubMed

Multifactorial pathways facilitate resilience among kindergarteners at risk for dyslexia: A longitudinal behavioral and neuroimaging study.

2021

Developmental science

Zuk J, Dunstan J, Norton E, Yu X, Ozernov-Palchik O +4 more

Plain English
This study looked at kindergarteners who are at risk for dyslexia but go on to read well, aiming to find out what factors helped them succeed. Researchers tracked these children until they reached second grade and found that those who read well tended to have higher socio-economic status (SES), better speech production, and stronger brain structure in a specific area associated with reading. This matters because understanding these protective factors could improve early screening methods and help allocate educational resources more effectively. Who this helps: This benefits educators and children at risk for dyslexia.

PubMed

The incidence of psoriasis in Chile: an analysis of the National Waiting List Repository.

2021

Clinical and experimental dermatology

Lecaros C, Dunstan J, Villena F, Ashcroft DM, Parisi R +4 more

Plain English
The study looked at how often psoriasis occurs in Chile by examining medical records from 2016 and 2017. Researchers found that about 73.3% of referrals from doctors to skin specialists were for new cases of psoriasis, with incidence rates at 22.1 and 22.7 per 100,000 people in those years. There were significant differences across regions, ranging from 0.75 cases in the Metropolitan area to 164.9 in the Aysen region, highlighting the need for better awareness and treatment options. Who this helps: Patients with psoriasis and healthcare providers in Chile.

PubMed

White matter in infancy is prospectively associated with language outcomes in kindergarten.

2021

Developmental cognitive neuroscience

Zuk J, Yu X, Sanfilippo J, Figuccio MJ, Dunstan J +6 more

Plain English
This study looked at how the brain's white matter in infants affects their language skills when they reach kindergarten. Researchers followed 40 children and found that specific areas of white matter are linked to better language abilities: for example, the connections in the left arcuate fasciculus were associated with awareness of sounds and vocabulary knowledge. These findings show that the structure of the brain in infancy is important for predicting a child's future language development. Who this helps: This helps parents and educators understand the importance of early brain development for language learning.

PubMed

Supporting the classification of patients in public hospitals in Chile by designing, deploying and validating a system based on natural language processing.

2021

BMC medical informatics and decision making

Villena F, Pérez J, Lagos R, Dunstan J

Plain English
This study focused on improving how patients in Chile are classified when seeking specialty consultations or surgeries. Researchers created a system that uses natural language processing to automatically determine if patient referrals are covered by a government health guarantee, achieving a high accuracy with an AUC score of 0.94. This is important because it reduces the risk of misclassifying patients, which can lead to longer wait times for necessary medical care, ultimately benefiting patients who need timely access to healthcare services. Who this helps: Patients in Chile needing specialty consultations or surgeries.

PubMed

Characteristics of COVID-19 in cancer patients: a cross-sectional study in Peru.

2021

Ecancermedicalscience

Payet E, Perez J, Sarria G, Neciosup S, Berrospi F +8 more

Plain English
This study looked at 100 cancer patients in Peru who caught COVID-19 to understand how the virus affected them. The results showed that 39% of these patients died, primarily due to severe breathing problems, with symptoms like cough (65%) and fever (57%) being common. Children's outcomes were better than adults', and most patients had serious health issues like anemia (67%) and elevated levels of inflammation markers. Who this helps: This information is valuable for doctors treating cancer patients with COVID-19.

PubMed

Correction to: Supporting the classifcation of patients in public hospitals in Chile by designing, deploying and validating a system based on natural language processing.

2021

BMC medical informatics and decision making

Villena F, Pérez J, Lagos R, Dunstan J

PubMed

Functional Connectivity in Infancy and Toddlerhood Predicts Long-Term Language and Preliteracy Outcomes.

2021

Cerebral cortex (New York, N.Y. : 1991)

Yu X, Ferradal SL, Sliva DD, Dunstan J, Carruthers C +8 more

Plain English
This study looked at how brain connections in babies and toddlers affect their language and reading skills later on in life. Researchers found that the way the brains of infants (under 18 months) were organized predicted their language abilities and early reading skills when they were around 6.5 years old, even after considering the influence of their environments. This is important because it shows that the brain's development in early childhood plays a critical role in later learning and communication skills. Who this helps: This helps parents and educators understand the importance of early brain development for language learning.

PubMed

A biomarker study in Peruvian males with breast cancer.

2021

World journal of clinical oncology

Castaneda CA, Castillo M, Bernabe LA, Sanchez J, Torres E +9 more

Plain English
This study focused on understanding the characteristics and potential biomarkers of breast cancer in 54 Peruvian males diagnosed over a 14-year period. Researchers found that most tumors were hormone receptor-positive, with 85.7% of cases being estrogen receptor positive and 85.3% positive for the androgen receptor. The study revealed that early-stage tumors and lower-grade cancers are linked to longer survival, while specific genetic markers like PIK3CA mutations were rare in this group. Who this helps: This research benefits male breast cancer patients and their doctors by providing insights into tumor characteristics and survival factors.

PubMed

[Construction of text resources for automatic identification of clinical information in unstructured narratives].

2021

Revista medica de Chile

Báez P, Villena F, Zúñiga K, Jones N, Fernández G +2 more

Plain English
This study focused on creating a system to automatically find important medical information in unstructured text from Chilean hospitals. Researchers gathered 150 patient referrals and trained three people to identify key medical terms, like diagnoses and medications. Initially, their accuracy measure (F1-Score) was 0.73, but after training, it improved to 0.87, showing that the training worked well. Who this helps: This benefits doctors and researchers by improving how they can access and use patient data.

PubMed

Predicting nationwide obesity from food sales using machine learning.

2020

Health informatics journal

Dunstan J, Aguirre M, Bastías M, Nau C, Glass TA +1 more

Plain English
This study looked at how to predict obesity rates in different countries by analyzing food sales data instead of conducting expensive surveys. The researchers used machine learning techniques on sales information from 79 countries and discovered that baked goods and flours, cheese, and carbonated drinks are key indicators of obesity rates. For example, their methods successfully predicted obesity levels with minimal error, which makes tracking this public health issue more efficient and cost-effective. Who this helps: Patients, public health officials, and policymakers.

PubMed

Effects of a sugar-sweetened beverage tax on prices and affordability of soft drinks in Chile: A time series analysis.

2020

Social science & medicine (1982)

Cuadrado C, Dunstan J, Silva-Illanes N, Mirelman AJ, Nakamura R +1 more

Plain English
This study looked at how a new tax on sugary drinks in Chile affected their prices and how affordable they were for consumers. After the tax was implemented in 2014, prices for carbonated drinks went up by 5.6%, and there was a noticeable reduction in affordability for carbonates, concentrates, and bottled water. This is important because it shows that while the tax aimed to reduce soda consumption, it also led to price increases that could make these drinks less affordable than before. Who this helps: This helps patients, especially those trying to manage their weight and consumption of sugary beverages.

PubMed

On the Construction of Multilingual Corpora for Clinical Text Mining.

2020

Studies in health technology and informatics

Villena F, Eisenmann U, Knaup P, Dunstan J, Ganzinger M

Plain English
This study focused on creating language-specific collections of medical text data to improve the analysis of healthcare information. Researchers successfully collected and organized 93,969 medical articles in English, German, and Spanish, totaling over 83 million words. This matters because it provides better resources for analyzing medical texts in multiple languages, enhancing the development of tools that can understand and process healthcare data more effectively. Who this helps: Patients and healthcare professionals who speak different languages.

PubMed

Earlier Breast Cancer Detection in Peru: Establishing a Comprehensive Program in an Underserved Region.

2020

Journal of the American College of Radiology : JACR

Matsumoto MM, Widemon S, Farfán G, Vidaurre T, Dunstan J +6 more

PubMed

Relationship between tumor-associated immune infiltrate and p16 staining over clinicopathological features in acral lentiginous melanoma.

2019

Clinical & translational oncology : official publication of the Federation of Spanish Oncology Societies and of the National Cancer Institute of Mexico

Castaneda CA, Castillo M, Torres-Cabala C, Bernabe LA, Casavilca S +11 more

Plain English
This study looked at how certain immune cells in tumors and the presence of a protein called p16 affect the outcomes for patients with a type of skin cancer known as acral lentiginous melanoma (ALM). Researchers analyzed 148 tissue samples and found that certain factors—like having less p16 protein (less than 10%) and thinner tumors—were linked to longer survival rates. Specifically, patients with thicker tumors or positive lymph nodes had poorer outcomes, with 54.2% showing lymph node involvement. Who this helps: This research benefits patients with acral lentiginous melanoma by highlighting factors that can influence their prognosis.

PubMed

Prolonged wait time is associated with increased mortality for Chilean waiting list patients with non-prioritized conditions.

2019

BMC public health

Martinez DA, Zhang H, Bastias M, Feijoo F, Hinson J +4 more

Plain English
This study looked at how long patients had to wait for medical care in Chile and how those wait times affected their chances of dying within two years. Out of nearly one million patients, 3.4% (33,546 people) died after being placed on waiting lists, and longer wait times were linked to higher death rates, especially among older men and patients needing cancer treatment. The findings show that reducing wait times could save lives and improve health outcomes for those in need of medical care. Who this helps: This helps patients waiting for treatment, particularly those with serious health conditions.

PubMed

Level of tumor-infiltrating lymphocytes and density of infiltrating immune cells in different malignancies.

2019

Biomarkers in medicine

Castaneda CA, Castillo M, Aliaga K, Bernabe LA, Casavilca S +20 more

Plain English
This study looked at immune cells called tumor-infiltrating lymphocytes (TIL) in samples from patients with different types of cancer, including gastric and cervical cancers. It found that higher levels of TIL were linked to better survival rates, especially in specific areas of the tumors. Notably, cutaneous melanoma showed the highest levels of immune cells. Understanding TIL levels can help doctors make better predictions about patient outcomes across various cancers. Who this helps: This benefits patients by providing insights that can improve treatment planning and prognosis.

PubMed

[Automatic keyword retrieval from clinical texts: an application of natural language processing to massive data of Chilean suspected diagnosis].

2019

Revista medica de Chile

Villena F, Dunstan J

Plain English
This research studied how to better analyze large amounts of medical text data from Chile, specifically looking at diagnostic suspicions for conditions not covered by a national health plan. The analysis included 2,592,925 diagnostic suspicions from health records collected between 2008 and 2018. The researchers developed a method to identify the most important keywords related to these diagnoses, making the data easier to understand and use for healthcare management and research. Who this helps: This helps doctors and researchers improve patient care and health system planning.

PubMed

Clinicopathological predictors of long-term benefit in breast cancer treated with neoadjuvant chemotherapy.

2018

World journal of clinical oncology

Galvez M, Castaneda CA, Sanchez J, Castillo M, Rebaza LP +8 more

Plain English
This study looked at 435 breast cancer patients who received chemotherapy before surgery to see how certain factors affected their survival. It found that only 11% of patients achieved a complete response to the treatment (meaning no detectable cancer in the tissue), but those with higher levels of lymphocytes in their tumors had better outcomes. The study highlights that some patient groups, especially those with certain types of breast cancer, may see better long-term benefits from treatment. Who this helps: This research helps doctors make better treatment decisions for breast cancer patients.

PubMed

Evaluating the 2014 sugar-sweetened beverage tax in Chile: An observational study in urban areas.

2018

PLoS medicine

Nakamura R, Mirelman AJ, Cuadrado C, Silva-Illanes N, Dunstan J +1 more

Plain English
The study looked at the effects of a sugar tax on soft drinks in Chile that started in October 2014. It found that people bought 21.6% less of the most sugary drinks after the tax was implemented, particularly among wealthier households. This is important because it shows how taxes on unhealthy foods can help reduce sugary drink purchases, which may lead to better health outcomes over time. Who this helps: This helps patients and health advocates aiming to reduce sugar consumption and improve public health.

PubMed

Critical review of axillary recurrence in early breast cancer.

2018

Critical reviews in oncology/hematology

Castaneda CA, Rebaza P, Castillo M, Gomez HL, De La Cruz M +5 more

Plain English
This study looked at how often early breast cancer comes back in the armpit area after specific surgeries. It found that about 2% of patients who had surgery to remove lymph nodes experienced this recurrence, which negatively affects their chances of survival. New methods like Sentinel Lymph Node Biopsy are being used to manage treatment, showing that they can safely reduce the risk of this recurrence, especially in specific patient groups. Who this helps: This helps breast cancer patients and their doctors in planning more effective treatments.

PubMed

Associations of,,, and LINE-1 methylation, measured in saliva, with waist circumference, body mass index, and percent body fat in mid-childhood.

2017

Clinical epigenetics

Dunstan J, Bressler JP, Moran TH, Pollak JS, Hirsch AG +3 more

Plain English
This study looked at 431 adolescents aged 10 to 15 to understand how certain genetic changes, measured in saliva, relate to body weight and fat. The researchers found that in boys who were obese, higher levels of methylation in specific genes were linked to lower body mass index, waist size, and body fat percentage. This is important because it highlights a new way to study obesity in young people using saliva, which could help identify genetic risk factors for obesity. Who this helps: This helps doctors and researchers understand obesity risks in children.

PubMed

Tumor infiltrating lymphocytes in acral lentiginous melanoma: a study of a large cohort of cases from Latin America.

2017

Clinical & translational oncology : official publication of the Federation of Spanish Oncology Societies and of the National Cancer Institute of Mexico

Castaneda CA, Torres-Cabala C, Castillo M, Villegas V, Casavilca S +9 more

Plain English
Researchers studied acral lentiginous melanoma (ALM), a type of skin cancer common in non-Caucasian populations, to understand how tumor-infiltrating lymphocytes (TILs) affect patient outcomes. They examined 537 ALM cases and found that lower levels of TIL were linked to larger tumors and poorer survival rates, with a median overall survival of just 5.3 years for early-stage ALM patients, compared to 9.2 years for those with melanoma in other limbs. This research highlights the aggressive nature of ALM and the importance of TIL levels in predicting outcomes for these patients. Who this helps: Patients with acral lentiginous melanoma and their doctors.

PubMed

[Accelerating cancer control innovations in Peru].

2016

Revista peruana de medicina experimental y salud publica

Santos-Ortiz C, Manrique J, Amorín E, Sarria G, Salazar M +5 more

Plain English
The study focused on innovative cancer prevention and treatment programs developed by the National Institute of Neoplastic Diseases (INEN) in Peru. From 2012 to 2015, they trained over 12,194 health professionals to improve skills in cancer prevention and diagnosis, especially for cervical cancer. This is important because better-trained healthcare workers can lead to improved cancer care and outcomes for patients across the country. Who this helps: This helps patients by ensuring they receive better cancer prevention and treatment from trained healthcare providers.

PubMed

Frequent Co-Authors

Fabián Villena Nadine Gaab Julio Abugattas Jennifer Zuk Carlos A Castaneda Miluska Castillo Henry L Gomez Xi Yu Joselyn Sanchez Henry Guerra

Physician data sourced from the NPPES NPI Registry . Publication data from PubMed . Plain-English summaries generated by AI. Not medical advice.