Teleassistance for remote home oxygen monitoring in neonates

Abstract

Objectives

To evaluate the feasibility and clinical impact of an IoT-based teleassistance system for remote monitoring of preterm neonates receiving home oxygen therapy.

Methods

A historically controlled cohort study was conducted in Manizales, Colombia, with three phases: (1) baseline data collection from neonates receiving traditional follow-up (n=32), (2) development of an Internet of Things (IoT)-based teleassistance system integrating physiological monitoring, cloud storage, and mobile applications, and (3) pilot testing in a prospective intervention cohort (n=25). The primary outcome was the duration of home oxygen therapy.

Results

Median duration of home oxygen therapy was reduced from 34 days (IQR 25–46) in the control group to 10 days (IQR 7–17) in the intervention group (p<0.001). Reliable communication was achieved through 4G and Wi-Fi, while LoRa performance was limited by geographic constraints.

Conclusions

IoT-based teleassistance was associated with a reduction in the duration of home oxygen therapy in preterm neonates and demonstrated feasibility for remote monitoring in resource-limited settings.

Keywords

home care services infant premature oxygen inhalation therapy remote monitoring retinopathy of prematurity telemedicine

Introduction

Premature neonates often require home-based supplemental oxygen therapy after discharge from neonatal intensive care units (NICUs) due to underdeveloped pulmonary systems or chronic conditions such as bronchopulmonary dysplasia and congenital heart disease.¹ This vulnerable population demands continuous and precise physiological monitoring to prevent complications, including retinopathy of prematurity (ROP), neonatal sepsis, and cardiovascular instability.^2,3

In Colombia, up to 40% of premature neonates born after 32 weeks receive home oxygen therapy.² However, current follow-up approaches—ranging from structured programs like kangaroo mother care to unstructured outpatient visits—lack standardized protocols for continuous data acquisition and timely therapeutic decision-making.

Non-invasive monitoring technologies, particularly photo-plethysmography (PPG), have appeared as promising tools in telemedicine and telecare environments. PPG sensors can measure oxygen saturation (SpO₂), heart rate (HR), and, with advanced modeling, estimate blood pressure (BP).^4–6 Accurate BP assessment in neonates is clinically significant, as their immature cardiovascular systems present unique hemodynamic characteristics such as high heart rate variability and reduced arterial compliance, which challenge direct adaptation of adult-based algorithms.^7,8 Despite advances in PPG-based BP estimation in adult populations,^9,10neonatal-specific models are scarce and often limited to NICU environments without integration into portable or IoT-ready platforms.¹¹

Addressing these gaps requires solutions that combine physiological modeling with embedded computing capable of operating under the constraints of homecare environments—limited connectivity, energy restrictions, and the need for caregiver-friendly interfaces, as shown in previous studies.¹² This platform integrates a wearable module with a MAX30102 PPG sensor for real-time acquisition of SpO₂, HR, and PPG-derived BP using a neonatal-adapted Windkessel model. Signal processing algorithms are optimized to manage neonatal motion artifacts and to use on resource-constrained devices, while communication is supported via 4G/LTE and Wi-Fi for reliable data transmission.^13–15

Objective: This study aims to evaluate the effectiveness and feasibility of an IoT-based teleassistance system, integrating PPG-based blood pressure estimation and multiparameter physiological monitoring, to reduce the duration of home oxygen therapy and improve clinical decision making for preterm neonates in Manizales, Colombia.

Materials and methods

This historically controlled cohort study was conducted in Manizales and Villamaría, Colombia, between November 2020 and May 2025, and included an embedded technological development phase to evaluate an Internet of Things (IoT)-based teleassistance system for neonatal home oxygen therapy. A three-phase methodological approach was implemented to evaluate the feasibility and effectiveness of an Internet of Things (IoT)-based teleassistance system for neonatal home oxygen therapy (Figure 1). The protocol was approved by a local institutional ethics committee (Act No. 087, 2019). Written informed consent was obtained from the parents or legal guardians of all participants prior to study initiation.

Phase 1: Baseline establishment (historical control cohort)

Between November 2020 and June 2021, prospective data were collected from neonates discharged with home oxygen therapy from the neonatal units of two local hospitals in Manizales, Colombia. Eligible participants were found through hospital discharge records and met the following inclusion criteria:

• Clinical indication for home oxygen therapy due to prematurity or chronic neonatal lung disease.

• Residence within the municipalities of Manizales or Villamaría.

• Discharge to home without access to structured telemonitoring.

Exclusion criteria included congenital heart disease or other major anomalies, incomplete medical records, or transfer to another institution. A total of 32 neonates met these criteria. Infants were discharged with home oxygen therapy when oxygen requirements were ≤0.5 L/min to maintain SpO₂ within the predetermined range (90–95%), and when they met additional criteria: weight gain of at least 20–30 grams/day outside the incubator for three days, no evidence of anemia (hematocrit and hemoglobin within the past week), physiological stability without acute illness (or with controlled chronic illness), and no apneas in the last week, following the guidelines of the Colombian Neonatology Association and the Colombian Pediatric Pulmonology Association.¹⁶ They had received supplemental oxygen therapy through unstructured follow-up programs or the kangaroo mother care program.

No teleassistance system was used in this cohort; follow-up was performed via scheduled outpatient visits, subject to appointment availability and authorization from healthcare providers.

Phase 2: Teleassistance system development

The intervention was developed as a comprehensive teleassistance service for remote physiological monitoring. The system included:

(a) A wearable monitoring module integrating sensors for oxygen saturation (SpO₂) and non-invasive blood pressure estimation using photoplethysmography (PPG);⁶

(b) IoT-based communication capabilities (4G/LTE and Wi-Fi, with experimental LoRa).

(d) A mobile application providing caregivers and healthcare professionals with alerts, patient data, and educational materials.

The system was designed for continuous monitoring and automated generation of visual and auditory alerts when physiological parameters exceeded predefined thresholds. The hospital care team received remote access to the monitoring dashboard to support timely clinical decision-making. The monitoring module was calibrated in neonatal units against standard clinical devices prior to home deployment.

The developed module measures oxygen saturation, heart rate, blood pressure, and temperature. Oxygen saturation, heart rate, and blood pressure estimation are derived from a MAX30102 optical plethysmography sensor (Maxim Integrated Products, Inc) processed via a Raspberry Pi 4 (Sunhokey Electronics Co) running custom Python algorithms. Visual and auditory alerts show abnormal values. Diagnostic reliability testing was conducted between the sensor’s readings and standard ICU monitors. Using Walter et al.’s method,¹⁷ a sample size of 616 observations was calculated to achieve 80% power and 95% confidence, with p₀: 0 and p₁: 0.1. The Pearson correlation coefficient for SpO₂ was 0.87.

Communication Network: IoT connectivity was enabled through 4G/LTE, Wi-Fi, and experimental LoRaWAN. Ideal LoRa coverage (Figure 2) was simulated for four routers; practical range was limited to 2 km line-of-sight with notable data loss. LoRa was only used when stable message reception under three minutes was confirmed. Mean reception interval for a 10-second send rate is shown in Table 1.

Table 1.

LoRa network performance with 10-seconds transmission interval.

Statistic	Values (seconds)
Mean	29.65
Standard deviation	18.79
Range	64.00
Minimum	10.00
Maximum	74.00

Information System: A web-based platform registers neonates, caregivers, treatment data, and physiological readings (Figure 3). An Android application (Figure 4) provides caregivers and clinicians with real-time alerts, educational materials, and visualizations.

Phase 3: Teleassistance pilot testing (intervention cohort)

From July 2024 to May 2025, a pilot study was conducted with neonates meeting the same inclusion and exclusion criteria as the historical control cohort, with the additional requirement of residence within an IoT network coverage area to ensure real-time data transmission. Recruitment took place at SES Hospital Universitario de Caldas and Clínica Ospedale before discharge, using consecutive sampling of all eligible neonates who met the predefined eligibility criteria. No formal matching procedure was performed between the intervention and historical control cohorts. Instead, participants were recruited consecutively until an intervention cohort of comparable size to the historical control group was achieved.

In the intervention cohort, home oxygen therapy was monitored continuously via the teleassistance module. Discontinuation criteria were standardized: maintenance of SpO₂ between 90% and 95% for 48 continuous hours on an auxiliary oxygen system. Oxygen weaning was performed by halving the oxygen flow at each step. Upon meeting these criteria, the teleassistance service was ended and the device retrieved. Clinical decisions were made following standardized institutional protocols supported by clinician judgment.

Interoperability testing was performed to integrate the system with existing hospital infrastructure and clinical workflows. Caregivers received on-site training at discharge and ongoing technical support during home use. Response to alerts and compliance with device usage were monitored throughout the study period.

Data analysis

The primary outcome variable was the duration of home oxygen therapy. Secondary outcomes included the corrected age at which home oxygen was stopped, the total home oxygen consumption and system connectivity performance.

A sample size of 23 infants per group was calculated to demonstrate a statistically significant difference of one week in the corrected gestational age at which oxygen was discontinued, with 80% power and a 0.05 significance level, using an average corrected age of 38.5 weeks for discontinuation of home oxygen in preterm infants with mild BPD as a reference. 18

Categorical variables were compared using Pearson chi-square or Fisher’s exact test. Continuous variables with normal distribution according to the Shapiro–Wilk test were compared using the student’s t-test; those without normal distribution were compared with the Mann–Whitney test. Finally, Cox regression and Kaplan–Meier analyses were performed by group (non-intervention vs intervention) and compared using the log-rank test. No multivariable adjustment was performed due to sample size limitations. Data were analyzed using Stata 16.1 (Stata Corp, TX, USA).

Results

Participant flow and recruitment period. A total of 32 neonates met the inclusion criteria for the historical control cohort (Phase 1). For the intervention cohort (Phase 3), 25 neonates were recruited, but 4 parents declined to take part. Recruitment for the control group took place between November 2020 and June 2021, while recruitment for the intervention group occurred between July 2024 and May 2025. Figure 5 summarizes participant recruitment, enrollment, and analysis according to TREND recommendations. Baseline demographic and clinical characteristics were comparable between cohorts. All participants in both cohorts met the same inclusion and exclusion criteria.

Table 2 describes the clinical characteristics of the patients. No significant differences were found between the groups in terms of gestational age, sex, birth weight, doses of pulmonary surfactant, length of hospital stay, or the duration of supplemental oxygen therapy at discharge.

Table 2.

Clinical characteristics of the studied patients by group.

	Non-intervention group		Intervention group		p-value
	n	%	n	%	p-value
Gestational age
Median	32		33		0.410^*
IQR	31 - 34		31 - 35
Gestational age group
<28	2	6.2	0	0.0	0.062^¥
28-31	10	31.2	2	9.5
32-33	9	28.1	6	28.6
34-36	11	34.5	13	61.9
Sex
Female	16	50.0	10	47.6	0.865^†
Male	16	50.0	11	52.4
Birth weight (grams)
Mean	1768.6		1885.4		0.429^‡
SD	552.6		471.8
95% CI	1569.3 – 1967.8		1670.6 – 2100.2
Hospital stay (days)					0.974^*
Median	20		18
IQR	12 – 28.5		14 - 22
Surfactant doses
0	17	54.8	11	52.4	1.000^¥
1	12	38.7	9	42.9
2	2	6.5	1	4.7
Days with oxygen before discharge
Median	18.5		18		0.622^*
IQR	10 – 27.5		14 - 21		0.622^*
Oxygen flow at discharge (L/min)
Median	0.062		0.062		1.000^*
IQR	0.062 – 0.125		0.062 – 0.125		1.000^*

*Mann-Whitney test.

^¥Fisher’s exact test.

^†Pearson’s chi-square test.

^‡Student’s T-test.

SD: standard deviation.

95% CI: 95% confidence interval.

IQR: interquartile range.

Table 3 shows the outcomes according to the group (non-intervention vs. intervention). The duration of home oxygen therapy was shorter in the intervention group (median 10 days, IQR 7-17 days, minimum 4 and maximum 56 days) compared to the non-intervention group (median 34 days, IQR 25-46 days, minimum 9 and maximum 127 days) (p<0.001). Additionally, the corrected age at which home oxygen was stopped was lower in the intervention group (median 36 weeks, IQR 34-36 weeks) compared to the non-intervention group (median 37 weeks, IQR 36-40 weeks, p<0.013). Similarly, total home oxygen consumption was lower in the intervention group (median 870 liters, IQR 671-1092) compared to the non-intervention group (median 2916 liters, IQR 1098-5622.5, p<0.001).

Table 3.

Outcomes by group.

	Non-intervention group	Intervention group	p-value
Range of duration (days)
Minimum	9	4	NA
Maximum	127	56
Total days of home oxygen therapy
Median	34	10	<0.001^*
IQR	25 - 46	7 - 17	<0.001^*
Corrected age at oxygen discontinuation
Median	37	36	0.005^*
IQR	36 - 40	34 - 36
Total home oxygen consumption (liters)
Median	2916	870	<0.001^*
IQR	1098 – 5622.5	671 - 1092	<0.001^*

*Mann-Whitney test.

IQR: interquartile range.

NA: not applicable.

Figure 6 compares the duration of home oxygen therapy between the two groups (Mann Whitney p: <0.001), and Figure 7 shows the Kaplan-Meier curves for the duration of home oxygen therapy according to group (log-rank test p: <0.001). The hazard ratio was 4.29 (95%CI: 2.34-7.89).

Adverse events

No adverse events or safety incidents related to device use were reported in the intervention cohort.

Discussion

This study aimed to evaluate the feasibility and clinical impact of an IoT-based teleassistance system for remote monitoring of preterm neonates on home oxygen therapy. The primary finding was a meaningful reduction in the duration of oxygen therapy in the intervention cohort, with a median of 10 days compared to an estimated 34 days in the historical control group. This reduction is clinically relevant, as it ensures that neonates receive only the duration of supplemental oxygen needed. These results reinforce our first hypothesis that continuous remote monitoring through IoT-enabled systems can improve therapy duration while supporting clinical safety and improving care efficiency.

The corrected gestational age at which home oxygen was discontinued was 36 weeks in the intervention group, in contrast to the study by Charpak and colleagues in Bogotá,¹⁸ who found a corrected gestational age of 38.5 weeks at the time of discontinuation and an average of 101 days of supplemental oxygen from birth. Bogotá is found at 2600 meters above sea level (masl). It has been proven that at altitudes above 2100 masl, oxyhemoglobin saturation begins to decline rapidly, increasing the likelihood that preterm newborns will be discharged with home oxygen and that its duration will be longer.¹⁶ Manizales is situated at 2130 masl, an intermediate altitude according to the World Health Organization.¹⁹

When compared to earlier studies in similar contexts,^18,20 the reduction observed in our cohort is notable. While kangaroo mother care and other structured follow-up programs have shown benefits in neonatal outcomes, their ability to promptly stop oxygen therapy is still constrained by scheduling and authorization processes inherent in traditional care models. By enabling real-time physiological data acquisition and automated alerts, our system supported timely clinical decision-making and facilitated earlier and safer discontinuation of oxygen therapy.

The present findings should also be interpreted within the broader context of telemedicine and remote patient monitoring. A growing body of evidence suggests that digital health interventions can improve access to specialized care, support continuity of care after hospital discharge, and facilitate timely clinical decision-making through the availability of real-time patient data. Telemonitoring has demonstrated benefits across multiple clinical domains, including chronic disease management, cardiovascular care, respiratory disorders, and pediatric follow-up programs. In neonatal care, however, the implementation of home-based telemonitoring systems remains comparatively limited. Our study contributes to this emerging field by demonstrating the feasibility of integrating physiological monitoring, IoT connectivity, and clinical teleassistance into a single platform for managing home oxygen therapy in preterm neonates. These findings support the growing role of digital health technologies in extending specialized neonatal care beyond the hospital setting and into the home environment.

Our results on the reduction in the duration of home oxygen therapy are consistent with those obtained by Rhein and colleagues²¹ in the United States, in their controlled clinical trial evaluating whether analysis of home oximetry data compared to standard care would reduce the duration of home oxygen therapy. Patients in the control group had an average of 100.1 days with oxygen, compared to 78.1 days in the intervention group. Their study was conducted in nine academic centers at altitudes ranging from 18 to 298 meters above sea level, but with extremely preterm newborns (average 26.8 weeks) and extremely low birth weight (average 938 grams), which differs from our population (33 weeks and 1885 grams), explaining the differences in the duration of home oxygen therapy.

The mechanism underlying these results likely involves a combination of continuous monitoring, caregiver engagement through the mobile application, and the availability of accurate, real-time data for clinicians. Continuous access to oxygen saturation and other physiological parameters allowed the healthcare team to identify stabilization earlier, reducing unnecessary days on supplemental oxygen. Furthermore, integrating photoplethysmography (PPG)-based blood pressure estimation⁶ within the teleassistance module provided additional hemodynamic insights, which may contribute to more comprehensive neonatal monitoring. While photoplethysmography has been widely studied in clinical and adult monitoring settings, its application in neonatal home-based monitoring stays limited. To our knowledge, few systems have integrated PPG-derived parameters into teleassistance platforms for remote monitoring of preterm neonates, highlighting the novelty of our approach in this context.

No significant differences were found about birth weight or oxygen flow at discharge, unlike the study by Yeh and colleagues,²² who found that patients with oxygen flow >0.5 L/min at discharge had longer durations of home oxygen therapy.

A key strength of this study is the real-world implementation of a digital health solution in a resource-limited setting with challenging geographical conditions. The teleassistance module, supported by a hybrid communication infrastructure (4G/LTE, Wi-Fi, and experimental LoRaWAN), functioned reliably in most households, with 4G proving to be the most consistent communication method. This integration of IoT and mobile health technologies aligns with global trends in digital transformation of healthcare and proves the potential for scalability. Experimental LoRaWAN testing revealed significant data loss and transmission delays under real-world conditions. With a transmission interval of 10 seconds, approximately 33% of transmitted messages were successfully received, and delays of up to 160 seconds were seen. These limitations reduced its suitability for reliable real-time monitoring, leading to the prioritization of 4G and Wi-Fi connectivity.

However, several limitations should be acknowledged. First, the nonrandomized, historically controlled design introduces the potential for selection bias and confounding, making it difficult to definitively attribute the observed differences solely to the telemonitoring intervention. Although no major changes in national guidelines for home oxygen withdrawal occurred during the study period and baseline demographic and clinical characteristics were comparable between cohorts, residual confounding related to unmeasured temporal changes in clinical practice cannot be completely excluded. Second, the sample size in the intervention cohort was relatively small, and recruitment was limited to households within IoT network coverage, which may affect the generalizability of the findings. Third, long-term outcomes beyond oxygen therapy discontinuation, such as respiratory readmissions or late-onset complications, were not assessed. Finally, LoRa coverage was substantially lower than theoretical estimates due to the mountainous geography of Manizales, which restricted its practical deployment and limited its suitability for reliable real-time monitoring.

Despite these limitations, the findings have important implications for neonatal care in similar settings. The system could be integrated into national or regional health programs to support structured, technology-enabled home follow-up, potentially reducing hospital readmissions, optimizing resource use, and improving caregiver confidence. Further studies should aim to confirm these results in larger, multicenter cohorts, explore the cost-effectiveness of the intervention, and assess its integration with electronic health record systems.

Conclusion

The IoT-based teleassistance service presented in this study offers a feasible and effective approach for remotely monitoring preterm neonates receiving home oxygen therapy. The intervention was associated with a reduction in the duration of home oxygen therapy, from a median of 34 days in the historical control cohort to 10 days in the intervention group, while supporting safe and evidence-based clinical decision-making. Continuous access to real-time physiological data, combined with automated alerts, facilitated earlier identification of clinical stability and timely discontinuation of oxygen therapy. These findings suggest that teleassistance systems may improve the efficiency of neonatal follow-up programs and help reduce the risks associated with prolonged oxygen exposure in this vulnerable population.

Beyond the immediate clinical benefits observed, the proposed teleassistance approach demonstrates the potential of IoT-enabled remote monitoring to support broader digital transformation strategies in neonatal care. From a technological perspective, the platform demonstrated the feasibility of integrating physiological monitoring, cloud-based information systems, and remote communication technologies into a unified teleassistance ecosystem capable of supporting home-based neonatal care. Hybrid connectivity based on 4G/LTE and Wi-Fi provided reliable communication, whereas LoRaWAN performance was limited by the mountainous geography of the study area, highlighting the importance of context-specific connectivity strategies in digital health deployments. The scalability and adaptability of the system make it particularly relevant for low- and middle-income countries where access to specialized follow-up and continuous physiological monitoring may be limited. Future research should evaluate long-term clinical outcomes, healthcare utilization, cost-effectiveness, implementation at larger scales, and integration with routine neonatal care pathways and electronic health records. These findings support further evaluation of teleassistance technologies as a strategy to improve access, continuity of care, and clinical outcomes for vulnerable neonatal populations.

Footnotes

Acknowledgments

The authors would like to thank Universidad Nacional de Colombia – Manizales campus, Universidad Autónoma de Manizales, S.E.S Hospital Universitario de Caldas, Clínica Ospedale, and the Ministerio de Ciencia, Tecnología e Innovación for their support during the project.

ORCID iD

Santiago Murillo-Rendón

Ethics considerations

The study was conducted following the Declaration of Helsinki and approved by the Bioethics Committee of the Universidad Autónoma de Manizales (approval record No. 087 of 2019). The project was classified as involving more than minimal risk due to the inclusion of a vulnerable population, by Resolution 8430 of 1993 on research involving human subjects in Colombia.

Consent to participate

Written informed consent was obtained from the parents or legal guardians of all neonates before enrollment in the study.

Author contributions

Conceptualization: Santiago Murillo-Rendón; Isabel C. Echeverri-Ocampo; Carolina Márquez-Narváez; Fernando Arango-Gómez; Belarmino Segura Giraldo; Camilo Salgado Jiménez; Oscar Julián López-Uribe. Methodology: Santiago Murillo-Rendón; Isabel C. Echeverri-Ocampo; Belarmino Segura Giraldo. Data curation: Carolina Márquez-Narváez; Fernando Arango-Gómez. Formal analysis: Santiago Murillo-Rendón; Camilo Salgado Jiménez. Writing—original draft: Santiago Murillo-Rendón; Isabel C. Echeverri-Ocampo. Writing—review and editing: All authors. All authors have read and agreed to the published version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Ministerio de Ciencia, Tecnología e Innovación de Colombia (MINCIENCIAS), project No. 121984468173, “Teleasistencia para Monitoreo de Oxígeno Domiciliario en Neonatos”, under Announcement 844-2019: “Convocatoria pacto para la generación de nuevo conocimiento a través de proyectos de investigación”.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Figure statement

All figures included in this manuscript are original and were created by the authors.

Use of artificial intelligence tools

The authors used artificial intelligence (AI) tools to support language editing and improvement of clarity and coherence of the manuscript. All scientific content, analysis, and conclusions were developed and confirmed by the authors.

Appendix

References

Hayes

Wilson

Krivchenia

, et al. Home oxygen therapy for children: an official American Thoracic Society clinical practice guideline. Am J Respir Crit Care Med 2019; 199: e5–e23. https://doi.org/10.1164/rccm.201812-2276ST

Pulido

Vargas

. Cuidado al recién nacido con oxígeno domiciliario [PhD thesis]. Universidad Nacional de Colombia, 2012.

Pezzotti

Torres Rodríguez

. Enfermedad pulmonar crónica del recién nacido: a 50 años del inicio de la ventilación mecánica en la etapa neonatal. Rev Esp Med Quir 2015; 21: 312–320.

Elgendi

Fletcher

Liang

, et al. The use of photoplethysmography for assessing hypertension. NPJ Digit Med 2019; 2: 60. https://doi.org/10.1038/s41746-019-0136-7

Allen

Zheng

Kyriacou

, et al. Photoplethysmography (PPG): state-of-the-art methods and applications. Physiol Meas 2021; 42: 100301. https://doi.org/10.1088/1361-6579/ac2d82

Jiménez

Echeverri-Ocampo

Segura Giraldo

, et al. Photoplethysmography-based blood pressure calculation for neonatal telecare in an IoT environment. Electronics 2025; 14: 3132. https://doi.org/10.3390/electronics14153132

Baker

Yogavijayan

Kandasamy

. Towards non-invasive and continuous blood pressure monitoring in neonatal intensive care using artificial intelligence: a narrative review. Healthcare (Basel) 2023; 11(24): 3107. https://doi.org/10.3390/healthcare11243107

Rao

Eskandar-Afshari

Weiner

, et al. Clinical study of continuous non-invasive blood pressure monitoring in neonates. Sensors (Basel) 2023; 23: 3690. https://doi.org/10.3390/s23073690

Ding

Zhang

Liu

, et al. Continuous cuffless blood pressure estimation using pulse transit time and photoplethysmogram intensity ratio. IEEE Trans Biomed Eng 2016; 63: 964–972. https://doi.org/10.1109/TBME.2015.2480679

10.

Huynh

Jafari

Chung

. Noninvasive cuffless blood pressure estimation using pulse transit time and impedance plethysmography. IEEE Trans Biomed Eng 2019; 66: 967–976. https://doi.org/10.1109/TBME.2018.2865751

11.

Krbec

Zhang

Chityat

, et al. Emerging innovations in neonatal monitoring: a comprehensive review of progress and potential for non-contact technologies. Front Pediatr 2024; 12: 1442753. https://doi.org/10.3389/fped.2024.1442753

12.

Amaya-Carrascal

Vera-Betancourt

Marquez-Narvaez

, et al. Development of an information system for the monitoring of physiological variables and information storage of neonates on domiciliary oxygen. In: Castillo Ossa

Isaza

Cardona

, et al. (eds). Trends in Sustainable Smart Cities and Territories. Springer, 2023, pp. 219–230. https://doi.org/10.1007/978-3-031-36957-5_19

13.

Vadillo

Martín

Malagón

, et al. The role of smart homes in intelligent homecare and healthcare environments. In: Dobre

Mavromoustakis

Garcia

, et al. (eds). Ambient Assisted Living and Enhanced Living Environments. Elsevier, 2017, pp. 345–394.

14.

Cowan

Rogers

Najafi

, et al. Electronic assistive technology. In: Taktak

Ganney

Long

, et al. (eds). Clinical Engineering. Elsevier, 2014, pp. 359–388.

15.

Turner

. The ecology of home sensor networks for telecare. In: Saito

Menga

(eds). Ecological Design of Smart Home Networks. Elsevier, 2015, pp. 51–66.

16.

Colombian Association of Neonatology (ASCON), Colombian Association of Pediatric Pulmonology (ACNP) . Expert consensus, https://ascon.org.co/wp-content/uploads/2021/08/08.2021-Consenso-Neumología-y-Neonatología.pdf Available from: [cited 2025 Jun 9].

17.

Walter

Eliasziw

Donner

. Sample size and optimal designs for reliability studies. Stat Med 1998; 17(1): 101–110. https://doi.org/10.1002/(sici)1097-0258(19980115)17:1<101::aid-sim727>3.0.co;2-e

18.

Charpak

Ruiz

Motta

. The clinical course and one year forecast for a cohort of premature infants who were discharged with home oxygen in Bogota, Colombia. Rev Salud Publica (Bogota) 2012; 14: 102–115. https://doi.org/10.1590/s0124-00642012000100009

19.

Crocker

Hossen

Goodman

, et al. Effects of high altitude on respiratory rate and oxygen saturation reference values in healthy infants and children younger than 2 years in four countries: a cross-sectional study. Lancet Glob Health 2020; 8: e362–e373. https://doi.org/10.1016/S2214-109X(19)30543-1

20.

Pallás-Alonso

. Need for home oxygen in premature infants living in the city of Bogotá. Pediatr Pulmonol 2021; 56: 1838–1840. https://doi.org/10.1002/ppul.25376

21.

Rhein

White

Simoneau

, et al. Transmitted home oximetry and duration of home oxygen in premature infants. Pediatrics 2020; 146: e20200079. https://doi.org/10.1542/peds.2020-0079

22.

Yeh

McGrath-Morrow

Collaco

. Oxygen weaning after hospital discharge in children with bronchopulmonary dysplasia. Pediatr Pulmonol 2016; 51: 1206–1211. https://doi.org/10.1002/ppul.23442