Wearable Sensors and Telemedicine Strategies for Cystic Fibrosis Patients

Introduction

Telemedicine is an emerging tool that enables clinicians to provide direct patient care at a distance by leveraging telecommunication information technology [1]. The recent worldwide spread of the COVID-19 pandemic has highlighted the necessity to develop reliable monitoring devices capable of analyzing physio pathological parameters directly at home. This necessity is not only related to patients with COVID-19 disease, but also to all those patients with chronic diseases that show difficulties to be followed up and treated in hospital, due to risk of infections during hospitalization and difficulties to reach the appropriate care centers. To this end, wearable sensors used as medical devices that can continuously collect data from the human body are becoming more and more widespread. Their high flexibility, together with the non-invasive measures and simple way of utilization for patients leads to possible applications as physical activity monitoring and personalized healthcare [2]. Cystic Fibrosis (CF) is a congenital disease which manifestations appears since intrauterine life and generally at the birth. It is inherited as an autosomal recessive manner. CF is characterized by chronic and progressive symptoms with a multiorgan localization [3].

Cystic Fibrosis Conductance Regulator (CFTR) is the diseasegene of CF, located on Chromosome 7, encoding a protein largely expressed in epithelial cells, which has the role of regulating the chloride efflux across the cellular membrane. About 2103 mutations are described in CFTR gene leading to protein degradation, protein misfunctions or protein mis localization [4]. Protein defects or insufficient CFTR transcript amounts are the cause of an alteration of the flow of sodium and chloride ions through cellular membrane, resulting in dense secretions accumulation obstructing epithelial ducts. A progressive respiratory system impairment with chronic inflammation and specific bacterial colonization is the main cause of death about CF patients [5]. Psychological and social impairments as well as depression and anxiety symptoms are described in adults and adolescents with CF [6] and reflects the severity and the complexity of disease management.

In the last decades, a multidisciplinary approach for therapies have significantly improved both the life quality and life expectancy in CF patients. Frequent disease exacerbations are strictly associated with a diffuse sedentary lifestyle among CF patients, partly determined by impaired muscle function and malnutrition and the lung disease that led to a reduced exercise tolerance [7]. Due to the complexity of each patient’s health status, exercise interventions could be continuously monitored, but also associated to mucus clearance techniques. Regular exercise is highly recommended in the care guidelines for CF patients, for the capacity of reduce mucus stagnation enhancing pulmonary functions, but also because it can produce an overall health state reducing anxiety and depression episodes and increasing sleep quality [8].

Last Advances in Telemedicine for Cystic Fibrosis

Recently, the COVID-19 pandemic has determined an increase of telemedicine and tele-health interventions for CF patients. Wearable activity monitors have been increasingly adopted to record data related to physical activity, such as heart rate, step count, distance traveled, elevation climbed and estimated energy expenditure, but also to monitor the variations of the quantity of electrolytes in sweat. Due to easily accessible collection and electrolytes composition, human sweat can reflect individual’s physiological state, and thus represents an attractive target for wearable sweet sensors [9,10]. Tomlinson and colleagues have observed successful interventions in monitoring via Skype calls exercise training of nine CF patients [11]. Wrist-worn activity monitors, using optical sensing technology to detect variation of blood volume in microvascular tissue, had controversial accuracy in literature if compared with golden standard technique for detection of heart rate during high intensity exercise [12,13]. Other studies had observed accuracy and low cost of technology for chest straps heart rate monitors worn during physical activity in healthy population [14] and CF population [15].

Another example of telemedicine is provided by the increasing number of programs based on imaging with digital fundus cameras and remote interpretation, which has made screening for retinopathy large populations associated with diabetes possible [16]. The advances in Artificial Intelligence (AI) and telemedicine have the potential to increase the accuracy, reduce the costs and extend the reach of screening, in the same way of glaucoma screening, using optic disc photographs and/or optical coherence tomography. This could indeed represent a cost-effective way to face glaucoma in populated nations like China and India, where most glaucoma cases still to remain undiagnosed [17]. Since physical activity plays an important role for the improving the quality of life for several chronic diseases, the development of Machine Learning methods capable of classifying human physical activity through data measured by on-body sensors such as accelerometers [18], heart rate monitors [19], a mix of accelerometers, cameras, GPS location and other sensors [20], becomes important in the determination of the appropriate physical activity needed to reduce the risks associated with chronic illness.

As far as CF is concerned, semi-automated devices which use audio and Electromygraphy (EMG) signals for cough detection were developed by [21]. Moreover, approaches based on AI techniques offer the possibility to correlate health state indicators and the type of physiotherapy exercise to be followed [22]. Artificial intelligence represents therefore a useful tool in determining the most appropriate physiotherapy. Not only, Zucker et al. Have recently shown that deep convolutional neural networks also facilitate Brasfield scoring of chest radiographies in patients with CF and thus the use of AI could also lead to important diagnostic improvements [23].

Conclusion

In conclusion, the use of AI can provide insights not only in the development of guidelines for choosing the best physical activity for CF patients, but it could also open new fields in diagnostics, enabling recognition of CF symptoms.

References

• Bashshur RL, Reardon TG, Shannon GW (2000) Telemedicine: a new health care delivery system. Annu Rev Public Health 21: 613-37.
• Gao W, Hiroki O, Daisuke K, Kuniharu T, Ali J (2019) Flexible electronics toward wearable sensing. Acc Chem Res 52(3): 523-533.
• Batten J (1965) Cystic fibrosis: A review. Br J Dis Chest 59: 1-9.
• Schneider Futschik EK (2019) Beyond cystic fibrosis transmembrane conductance regulator therapy: A perspective on gene therapy and small molecule treatment for cystic fibrosis. Gene Ther 26: 354-362.
• Moore JE, Mastoridis P (2017) Clinical implications of Pseudomonas aeruginosa location in the lungs of patients with cystic fibrosis. J Clin Pharm Ther 42(3): 259-267.
• Georgiopoulos AM (2019) Starting early: An international focus on mental health in children living with cystic fibrosis. Pediatr Pulmonol 55(2): 267-269.
• Burtin C, Hebestreit H (2015) Rehabilitation in Patients with Chronic Respiratory Disease Other than Chronic Obstructive Pulmonary Disease: Exercise and Physical Activity Interventions in Cystic Fibrosis and Non-Cystic Fibrosis Bronchiectasis. Respiration 89(3): 181-189.
• Cox NS, Eldridge B, Rawlings S, Dreger J, Corda J, et al. (2019) A web-based intervention to promote physical activity in adolescents and young adults with cystic fibrosis: Protocol for a randomized controlled trial. BMC Pulm Med 19: 253-258.
• Bariya M, Nyein H Y, Jevey A (2018) Wearable sweat sensors. Nat Electron 1: 160-171.
• Koh A, Kang A, Xue Y, Lee S, Pielak RM, et al. (2016) A soft wearable microfluidic device for the capture, storage, and colorimetric sensing of sweat. Sci Transl Med 8: 366ra165.
• Tomlinson O W, Shelley J, Trott J, Bowhay B, Chauhan R, et al. (2020) The feasibility of online video calling to engage patients with cystic fibrosis in exercise training. J Telemed Telecare (6): 356-364.
• Dooley EE, Golaszewski NM, Bartholomew JB (2017) Estimating accuracy at exercise intensities: a comparative study of self-monitoring heart rate and physical activity wearable devices. JMIR M health Unhealth 5(3): e34.
• Jo E, Lewis K, Directo D, Kim MJ, Dolezal BA (2016) Validation of Biofeedback Wearables for Photoplethysmographic Heart Rate Tracking. J Sports Sci Med 15(3): 540-547.
• Gillinov S, Etiwy M, Wang R, Gordon B, Dermot P, et al. (2017) Variable accuracy of wearable heart rate monitors during aerobic exercise. Med Sci Sport Exercise 49(8): 1697-1703.
• Holmes H, Wood J, Jenkins S, Peta W, Dianne L, et al. (2013) Xbox Kinect^TM represents high intensity exercise for adults with Cystic Fibrosis. J Cyst Fibrosis 12(6): 604-608.
• Rajalakshmi R, Prathiba V, Arulmalar S, Usha M (2020) Review of retinal cameras for global coverage of diabetic retinopathy screening. Eye (Lond) 35: 162-172.
• Tan NYO, Friedman DS, Stalmans I, Ahmed IIK, Sng CCA (2020) Glaucoma screening: where are we and where do we need to go? Curr Opin Ophthalmol 31(2): 91-100.
• Mannini A, Sabatini AM (2010) Machine Learning Methods for Classifying Physical Acivity from On-Body Accelerometers. Sensors 10: 1154-1175.
• Reiss A and Didier S (2012) "Creating and benchmarking a new dataset for physical activity monitoring." PETRA 40: 1-8.
• Li RT, Scott RK, Michael JS, Sean A, Joseph S, et al. (2016) "Wearable performance devices in sports medicine." Sports health 8(1): 74-78.
• Hamutcu R, Francis J, Karakoc F, Bush A (2002) Objective monitoring of cough in children with cystic fibrosis. Pediatr Pulmonol 34(5): 331-335.
• Slavici T, Bogdan A (2013) Artificial intelligence techniques: an efficient new approach to challenge the assessment of complex clinical fields such as airway clearance techniques in patients with cystic fibrosis? J Rehabilitation Medicine 45(4): 397-402.
• Zucker EJ, Zachary AB, Matthew PL, Yekaterina S, Jayne MS, et al. (2020) Deep learning to automate Brasfield chest radiographic scoring for cystic fibrosis. J Cystic Fibrosis 19(1): 131-138.