Our article selected as Editor's Choice by MDPI’s Bioengineering
Our team has created a simple way to tell the difference between moments that cause anxiety and moments that don't, using portable devices.
In everyday life, we all experience both happy and sad moments, whether from personal situations, watching emotional videos, or listening to stirring music. As anxiety becomes more common and mental health resources are limited, our team has created a simple way to tell the difference between sad moments that cause anxiety and happy moments that don't. This method uses portable devices to help assess anxiety levels.
In March 2022, the World Health Organization (WHO) reported that the COVID-19 pandemic led to a 25% increase in cases of anxiety and depression around the world1. Sadly, this rise in mental health issues happened at the same time as major disruptions to mental health services, leaving many people without the help they need1. To get mental health care, the first step is assessment. Currently, this is done through questionnaires during a medical interview2. But with long waiting times for appointments and the possibility of biased self-assessments, this method has its flaws. Imagine if we could evaluate anxiety more quickly and objectively, like using a portable device to collect body signals.
Our Master's student, Florian Ritsert, worked with project lead Dr. Moe Elgendi, PhD student Valeria Galli, and Prof. Carlo Menon on a study that aimed to detect anxiety using a portable device. This device records heart (ECG) and breathing (RSP) signals through video. Our earlier research showed that the ECG and RSP signals are particularly helpful in identifying stress using different machine learning methods3,4. The current study demonstrated the potential to monitor anxiety by evaluating changes in breathing rates obtained from RSP signals compared to ECG. Additionally, this study discovered a new finding: a transitional phase between stress-induced and non-stress-induced states [6]. Pinpointing anxiety biomarkers is crucial for developing a machine-learning algorithm and, ultimately, the automatic detection of anxiety5. Portable and wearable devices could help gather, analyze, and interpret these biosignals, offering a quicker and more objective way to assess anxiety.
Doesn’t that sound promising? Bioengineering thought so, and selected the paper to be one of their "Editor's Choice Articles". See the full paper here: external page Heart and Breathing Rate Variations as Biomarkers for Anxiety Detection.
1. WHO News: external page COVID-19 pandemic triggers 25% increase in prevalence of anxiety and depression worldwide (who.int)
2. Ritsert, F.; Elgendi, M.; Galli, V.; Menon, C. external page Heart and Breathing Rate Variations as Biomarkers for Anxiety Detection. Bioengineering 2022, 9, 711. https://doi.org/10.3390/bioengineering9110711
3. Elgendi, M.; Menon, C. external page Assessing Anxiety Disorders Using Wearable Devices: Challenges and Future Directions. Brain Sci. 2019, 9, 50. https://doi.org/10.3390/brainsci9030050
4. M. Elgendi and C. Menon, "external page Machine Learning Ranks ECG as an Optimal Wearable Biosignal for Assessing Driving Stress," in IEEE Access, vol. 8, pp. 34362-34374, 2020, doi: 10.1109/ACCESS.2020.2974933.
5. Ancillon, L.; Elgendi, M.; Menon, C. external page Machine Learning for Anxiety Detection Using Biosignals: A Review. Diagnostics 2022, 12, 1794. https://doi.org/10.3390/diagnostics12081794
