Validation of a Mobile Health Device for Neonatal Jaundice Screening: A Cross-Sectional Study in a Resource-Limited Setting in Mexico

Abstract

Objective:

To evaluate the performance of Picterus Jaundice Pro (Picterus JP), a mobile health device for neonatal jaundice (NNJ) screening, in Mexican newborns.

Methodology:

A cross-sectional study was conducted from January 2023 to June 2024 at a hospital in a resource-limited setting in Mexico. The main outcomes were Picterus JP measurements and total serum bilirubin (TSB) levels.

Results:

A total of 177 term newborns, aged 1 to 14 days, were enrolled. Picterus JP showed a significant positive Spearman’s rho correlation with TSB (ρ = .68), sensitivity of 85.7% and specificity of 80.1% to detect TSB ≥ 250 using Picterus cut off value of 202 µmol/L. However, it tended to underestimate higher bilirubin levels.

Conclusion:

Picterus JP shows potential to be a useful tool for NNJ screening, particularly in resource-limited areas. Further validation across more diverse populations and clinical environments, alongside accuracy improvements, is necessary to enhance its utility and support wider implementation.

Registered in ClinicalTrials.gov (https://clinicaltrials.gov/study/NCT06276582)

Keywords

neonatal jaundice neonatal hyperbilirubinemia detection screening mobile health resource-limited settings low-resource settings

Introduction

Neonatal jaundice (NNJ), a clinical sign of hyperbilirubinemia, is a global health issue affecting over half of newborns within the first 48 to 72 hours after birth.¹ If undetected or untreated, it can cause permanent brain damage, leading to long-term disabilities or even death.^2
-4 While most NNJ cases are harmless and self-limiting, around 1 million of newborns annually develop severe NNJ, requiring close monitoring and treatment.⁵

Accurate detection of NNJ remains a significant challenge, especially in low- and middle-income countries (LMICs), other resource-limited settings, and non-hospital locations. The gold standard for diagnosing NNJ, the total serum bilirubin (TSB) test, requires laboratory facilities and is invasive. Screening devices like transcutaneous bilirubinometers (TcB), while offering high sensitivity for detecting hyperbilirubinemia (74%-100%),⁶ are often inaccessible due to their high cost, typically ranging from $3,000 to $6,000 USD.^7
-11 Consequently, detection frequently relies on visual assessment (VA), which is inaccurate and unreliable, contributing to a higher prevalence of severe NNJ in these settings compared to high-income countries.^1,12,13

The widespread adoption of smartphones has paved the way for mobile health (mHealth) tools, offering non-invasive and potentially affordable alternatives for NNJ screening. Several smartphone-based devices have shown promise in early identification of severe cases.^14
-19 Picterus Jaundice Pro (Picterus JP) is a mobile device developed to support NNJ screening and follow-up. It includes a smartphone app for capturing and uploading images, a calibration card for adjusting light conditions and camera performance, and a server for image storage and analysis. The device uses optical measurements of bilirubin levels from images of a newborn’s chest. Picterus JP has been validated in Norway and pilot studies in Mexico and the Philippines, demonstrating a positive correlation with TSB levels and superior accuracy compared to VA.^20
-23

Despite significant advancements, smartphone-based tools like Picterus JP still face limitations, including sensitivity to varying lighting conditions and skin pigmentation. These challenges have been addressed through ongoing improvements to the app algorithm and the calibration card. Unlike prior validations conducted in Mexico, Norway, and the Philippines, this study evaluates an updated Picterus JP with a newer version of the calibration card in a distinct resource-limited setting in Mexico.

With nearly 130 million inhabitants and a diverse healthcare system,^24,25 Mexico presents a unique context for this evaluation. This study focuses on the device’s accuracy as a screening tool for NNJ in newborns from a resource-limited setting, providing valuable insights into its practical implementation and potential to improve neonatal care in similar global contexts.

Material and Methods

Study Setting and Design

A cross-sectional study was conducted at Hospital General de Zona 1 (HGZ) in Oaxaca de Juarez, Mexico, from January 2023 to June 2024. Oaxaca, known for its ethnic, linguistic, and cultural diversity,²⁶ is one of the country’s poorer states. This is a second-level hospital under the Instituto Mexicano del Seguro Social (IMSS), providing specialized care such as general surgery, obstetric services ad neonatal care, handling approximately 2,000 births annually.

Participants and Sample Size

Participants were recruited from the maternity ward and emergency room. Inclusion criteria were term newborns (gestational age of 37 weeks or more), aged 1 to 14 days, with a birth weight of ≥2500 grams, and who required a blood test for any medical reason to minimize unnecessary punctures. Newborns with congenital diseases, those transferred to paediatric wards for reasons unrelated to NNJ treatment, or those who had undergone phototherapy for NNJ were excluded.

The sample size was determined based on sensitivity and specificity analysis, assuming a severe jaundice (TSB ≥ 250 µmol/L; 14.6 mg/dL) prevalence of 10%. A sample size of 120-200 newborns was required to achieve a minimum power of 80% for an expected sensitivity and specificity between 80% and 90%.²⁷ Thus, the minimum sample size was set at 150 participants.

Study Procedures

Medical staff from the paediatric ward, trained and supervised through regular visits by the first author, collected the data. After obtaining informed consent, background information was retrieved from medical records, including birth time and date, birth weight, and gestational age. The Fitzpatrick scale was used to assess parental skin types. This classification categorizes adult skin based on its reaction to sun exposure and includes six categories ranging from very light/pale to very dark brown tones.²⁸ Newborn skin tones were classified using the Neomar scale, a tool specifically developed for newborns. This scale assesses skin pigmentation in areas such as the chest, areola, and genitals, classifying skin tones into four groups, ranging from white/pink to brown/dark tones.²⁹

Visual Assessment

VA was performed using the Kramer scale, a clinical method for estimating the severity of jaundice in newborns. The scale divides the body into five zones, starting from the head and progressing downward (cephalo-caudal progression), as jaundice spreads with increasing bilirubin levels in the blood. A score from 1 to 5 is assigned based on the lowest zone where yellow discoloration is visible.³⁰

Image Capture

A clinical trial version of Picterus JP (2.7.3, Picterus AS, Norway) was installed on randomly selected smartphones based on their availability at the study site (Samsung A23, iPhone 11, and iPhone 15 Pro) to obtain bilirubin results. During the procedure, a calibration card was placed on the newborn’s chest, and six images were automatically captured – three with flash and three without. These images were stored for later analysis. To ensure unbiased results, medical staff collecting the data were blinded to the bilirubin levels obtained by Picterus JP, preventing any influence on standard neonatal care. A unique measurement ID was recorded to link clinical data with the digital images after the study was completed. Upon completion, bilirubin levels were calculated using the Picterus JP algorithm (Jaundice Image Estimates [JIE], 1.9.14, Picterus AS).

TSB Measurements

TSB levels were measured with a blood sample obtained by venous puncture on the back side of the newborn’s hand, within one hour of capturing the images. Samples were immediately processed at the hospital laboratory by a vanadate oxidation method using an Atellica® CH analyser (Siemens Healthineers AG, Germany).

Data Analysis

Descriptive statistics were used to report the participants’ clinical characteristics. Normality of the TSB, Picterus JP, and Kramer scale values was assessed using the Kolmogorov-Smirnov test. Since the data were non-normally distributed, Spearman’s rho correlation coefficient (ρ) was used to determine correlations between Picterus JP, the Kramer scale, and TSB measurements. A Bland-Altman (BA) analysis evaluated the agreement between Picterus JP and TSB, identifying potential biases and outliers. Given the non-normal distribution of the mean difference, both transformed and untransformed datasets were analysed, yielding comparable results. For clearer clinical interpretation, the BA plot with the original, untransformed data is presented. Receiver Operating Characteristic (ROC) curves were generated, and the area under the curve (AUC) was used to evaluate the sensitivity and specificity of detecting severe NNJ at different Picterus JP and Kramer scale cut-off levels. For this study, severe NNJ was defined as TSB ≥ 250 µmol/L (14.6 mg/dL) based on TcB thresholds commonly used to confirm the diagnoses with TSB, as recommended in several clinical guidelines for NNJ.^31
-33 Given that Picterus JP is designed as a screening device, this definition seemed appropriate. Data analysis was conducted using IBM SPSS Statistics version 29.0.1.0 and GraphPad Prism version 10.3.0.

Ethical Considerations and Consent to Participate

The study was approved by ethical committees in Norway and Mexico: the Regional Committee for Medical and Health Research Ethics (REK), reference number 519379, and the National Committee for Scientific Research, IMSS, reference number R-2022-785-061.

The study objectives were explained to the participants’ parents, who provided written informed consent prior to participation. Confidentiality and anonymity of the participants were ensured. All methods were carried out in accordance with relevant guidelines and regulations.

Results

A total of 177 newborns were recruited during the study period. Thirteen were excluded due to not meeting the inclusion criteria or technical issues such as poor image quality, leaving 164 participants for analysis. Among these, 28 participants (17%) were classified as having severe NNJ.

The descriptive statistics of the participants’ clinical characteristics, categorized by age at inclusion, are presented in Table 1. Birthweight and gestational age were comparable across the age groups. TSB levels tended to increase after 72 hours post-birth.

Table 1.

Descriptive Statistics of Clinical Characteristics of Participants, Categorized by Age at Inclusion.

	All ages N = 164		1-3 days n=108		>3 < 7 days n = 46		≥ 7 days n = 10
Variable	Mean ± SD	Range	Mean ± SD	Range	Mean ± SD	Range	Mean ± SD	Range
Birthweight (grams)	3189 ± 399	2500-4480	3180 ± 405	2500-4480	3196 ± 396	2540-4290	3201 ± 379	2520-3920
Gestational age (weeks)	38.7 ± 1.2	37-42	38.8 ± 1.2	37-42	38.7 ± 1.1	37-41	38.9 ± 1.3	37-41
TSB (µmol/L) (mg/dL)	187.3 ± 80.3 10.9 ± 4.7	58-511 3.4-29.9	158.1 ± 51.9 9.2 ± 3	77-376 4.5-22	245.1 ± 90.4 14.3 ± 5.2	74-511 4.3-29-9	235.4 ± 120.3 13.7 ± 7	58-431 3.4-25.2
Picterus JP (µmol/L) (mg/dL)	172.5 ± 57.1 10.1 ± 3.3	8-319 0.5-18.7	154.8 ± 47.29 ± 2.7	64-290 3.7-17	206.1 ± 57.212 ± 3.3	8-308 0.5-18	209 ± 72.1 12.2 ± 4.2	95-319 5.6-18.7

SD: Standard Deviation; TSB: Total Serum Bilirubin; µmol/L: Micromoles per Litre; mg/dL: Milligrams per Decilitre;.Picterus JP: Picterus Jaundice Pro.

The majority of mothers’ skin types were classified as Fitpatrick scale score III (Table 2). The skin type for fathers was less frequently available, as they are less likely to be present in the hospital. Most newborns’ skin colours were classified as light to medium dark according to the Neomar scale (Table 3).

Table 2.

Fitzpatrick Scale Scores for Parents’ Skin Type.

	Mother	Father
Score	n (%)	n (%)
II	27 (16.5)	3 (1.8)
III	85 (51.8)	4(2.4)
IV	49 (29.9)	13 (7.9)
V	2 (1.2)	3 (1.8)
Missing	1 (0.6)	141 (86.0)
Total	164 (100)	164 (100)

Table 3.

Neomar Scale Score for Participants’ Skin Tone.

Score	n	(%)
1	62	37.8
2	54	32.9
3	45	27.4
4	3	1.8
Total	164	100

A statistically significant positive correlation was found between TSB and Picterus JP measurements (Spearman’s rho ρ = .68, 95% Confidence interval [CI]: 0.59-0.76, p ≤ .001) (Figure 1). There was a significant positive correlation between TSB and the Kramer scale (ρ = .72, 95% CI: 0.63-0.79, p < .001).

Figure 1.

Scatter plot showing the correlation between TSB and Picterus JP.

A Bland-Altman analysis revealed a mean difference (bias) ± SD of −14.73 ± 59.2 µmol/L (−0.86 ± 3.5 mg/dL) between Picterus JP and TSB, with 95% limits of agreement ranging from −130.7 to 101.25 µmol/L (−7.64 to 5.93 mg/dL) (Figure 2). To further explore the performance of Picterus JP, the dataset was split into two groups based on TSB values: ≤250and >250 µmol/L. Separate Bland-Altman analyses were then performed for each group. For TSB values ≤250 µmol/L, the mean difference was 0.85 ± 45.6 µmol/L, while for TSB values >250 µmol/L, the mean difference was −90.3 ± 59.6 µmol/L (Supplemental Figure 1(a) and (b)). These results indicate that Picterus JP tends to underestimate TSB levels, particularly when values exceed 250 µmol/L.

Figure 2.

Bland-Altman analysis comparing Picterus JP and TSB values in Micromoles per litre (μmol/L).

ROC analysis to determine Picterus JP’s accuracy in detecting TSB ≥ 250 µmol/L (14.6 mg/dL) showed an AUC of 0.875 (Figure 3(a)). Using a Picterus JP cutoff of 202 µmol/L (11.9 mg/dL) based on the Youden index yielded a sensitivity of 85.7% (correctly identifying most cases), specificity of 80.1% (effectively ruling out cases without severe jaundice), a positive likelihood ratio (+LR) of 4.3, and a negative likelihood ratio (−LR) of 0.17.

Figure 3.

Picterus JP (a) and Kramer scale (b) ROC curve for detecting TSB ≥ 250 μmol/L. μmol/L: Micromoles/Litre; AUC: Area Under the Curve; CI: Confidence Intervals.

Similarly, ROC analysis for the Kramer scale’s accuracy in detecting TSB ≥ 250 µmol/L (14.6 mg/dL) showed an AUC of 0.907 (Figure 3(b)). A Kramer scale cutoff of 3, corresponding to 200 µmol/L (11.7 mg/dL), resulted in a sensitivity of 92.9% and a specificity of 85.6%.

Discussion

In this study, we evaluated the performance of Picterus JP as a screening tool for NNJ in Mexican newborns at a second-level hospital in a resource-limited setting. Our main findings indicate that Picterus JP has a strong correlation with TSB levels and demonstrates satisfactory accuracy in identifying newborns who may need further diagnosis or follow-up for NNJ. The +LR suggests that a newborn with severe jaundice is 4.3 times more likely to have a positive Picterus JP value (cut-off ≥ 202 μmol/L) compared to one without the condition. However, the device tends to underestimate high bilirubin levels, highlighting areas where improvements are needed. Unlike other reports,^13,34,35 the Kramer scale showed outstanding performance and accuracy in this study.

The mothers’ skin types, classified using the Fitzpatrick scale, were consistent with the expected population characteristics for this region, where types III to V (light to dark brown) are most common.³⁶ However, significant missing data on fathers’ skin types limited our ability to analyse this factor.

The Neomar scale was used to assess newborn skin tones for the first time by the data collection staff, but the results did not align with the expected distribution for the studied population. Nearly 40% of the infants were classified with a score of 1, representing the lightest skin tone, which does not reflect the anticipated variation. This discrepancy highlights the subjective nature of the Neomar scale and its limitations in providing an accurate estimation of actual newborn skin tones. Consequently, these data were not used for further analysis.

Our findings are partially consistent with other studies on smartphone-based technologies designed to estimate bilirubin levels by analysing newborn skin colour. For instance, the BiliCam app, tested on 530 newborns including 26.3% Hispano-Latino, reported a correlation of 0.90 with an overall sensitivity of 100%, specificity of 76%, and a +LR of 4.2 for detecting TSB levels of 291 μmol/L at a cutoff of 222 μmol/L.¹⁶ The BiliScan app from Shenzhen Beishen Healthcare in China showed sensitivity between 75% and 90% and a TSB correlation ranging from 0.54 to 0.82, mainly in Middle Eastern and Chinese populations. Variations in inclusion parameters and result analysis in BiliScan studies limit the generalizability of their findings.^17,18,37
-39

Our results show a lower correlation between Picterus JP and TSB compared to earlier studies from Norway and Mexico (r = 0.85 and 0.87, respectively),²¹ but are consistent with a study from the Philippines (r = 0.70).²² Although the correlation coefficient is commonly used to assess the relationship between two methods, it is not the most suitable for evaluating agreement.⁴⁰ Bland-Altman analysis provides a more appropriate measure by assessing both bias and variability, offering a clearer and more comprehensive picture of test agreement.⁴¹ Table 4 summarizes the results of this analysis and Picterus JP’s accuracy for detecting TSB ≥ 250 μmol/L across different settings.

Table 4.

Summary of Results From Studies Using Picterus JP.

Study site	N	Mean difference Picterus JP – TSB		95% limits of agreement		AUC	Sensitivity^#	Specificity^#	Picterus JPcut-off
Study site	N	μmol/L	mg/dL	μmol/L	mg/dL	AUC	Sensitivity^#	Specificity^#	Picterus JPcut-off
Norway²⁰	302	−0.2	−0.01	−83.9 to 83.5	−4.9 to 4.9	0.92	87	77	225
Norway²³	201	–9.7	-0.56	−89.9 to 70.6	−5.6 to 4.1	0.89	94.1	70.7	214
This study	164	−14.7	−0.86	−130.7 to 101.2	−7.6 to 5.9	0.87	85.7	80.1	202

µmol/L: Micromoles per litre; mg/dL: Milligrams per decilitre; AUC: Area Under the Curve.

To detect TSB ≥ 250 μmol/L.

This study, consistent with previous research on Mexican newborns, indicates that Picterus JP tends to underestimate TSB, especially for values above 250 μmol/L. This pattern mirrors the underestimation observed with TcB measurements,⁸ possibly due to both Picterus JP and TcB assessing extravascular bilirubin, which can be up to four times lower than blood bilirubin concentration.⁴²

Variations in Bland-Altman analysis and accuracy across studies may be attributed to several factors, including differences in image quality, bilirubin kinetics, skin thickness, blood flow, skin maturity, and melanin levels across different races and ethnicities.^43
-45 The use of different prototype versions of the calibration card in previous studies may have led to inconsistent adjustments for lighting conditions across them. Additionally, the diversity in smartphones and Picterus JP versions could contribute to differences. Variations in participant inclusion criteria, study procedures, and laboratory equipment⁴⁶ may also result in discrepancies in TSB values and study outcomes.

However, these factors do not fully account for the wider 95% limits of agreement observed in our study. While these limits suggest that Picterus JP may not be suitable as a standalone diagnostic tool or replacement for TSB tests, it can serve its intended purpose as an effective screening tool. When used with adjusted cut-off values, the device can assist healthcare providers in identifying newborns who require a TSB test for confirmation, thereby reducing unnecessary blood draws and allowing earlier detection of jaundice.

Visual assessment remains a common method for screening NNJ worldwide, but its accuracy and reliability are often questioned due to its subjectivity and dependence on the examiner’s skill.^13,47 It has generally been more effective at ruling out severe NNJ than detecting it. The outstanding performance of Kramer scale in this study is likely due to the expertise of the paediatric professionals involved in the assessment. These results emphasize the importance of thorough training in visual assessment to improve the identification of severe cases, especially where other diagnostic tools are limited or not available.

Strengths

Data collection was performed by three trained medical doctors, ensuring high-quality and consistent results. The use of the same laboratory equipment for blood tests minimized potential discrepancies in TSB results.⁴⁶ Although the Asian population was not included in this study, the similarity in skin types between Hispanic-Latino and Asian groups suggests that the device is likely to perform well across these populations, increasing global applicability.

Additionally, as the Mexican government expands mHealth technologies to enhance healthcare services,⁴⁸ the study’s findings support the integration of Picterus JP into the national healthcare system, potentially improving early detection of severe NNJ, especially in resource-limited areas.

Limitations

The study was conducted in a single general hospital, with 65% of participants being younger than 72 hours. Since bilirubin levels typically continue to rise after the first 72 hours of birth, the results may not fully generalize to older newborns.

Our sample was limited to newborns who required blood tests, avoiding unnecessary invasive procedures. Most of these tests were conducted for routine or minor concerns, such as maternal urinary tract infections, rather than for serious medical conditions. Although the prevalence of severe NNJ (TSB > 250 µmol/L) in our study is consistent with existing estimates for this condition, this sampling limitation should be considered when interpreting the study’s applicability to all newborns.

The selected smartphones were chosen based on availability and reflect commonly used models with different software systems. While the calibration card is designed to correct for variations in camera sensors, we cannot entirely rule out the possibility of bias in the study data due to smartphone variability.

The findings may also be more applicable to newborns with similar skin types to the study population, as Picterus JP’s performance has not been validated for infants with higher melanin content, limiting its generalizability to populations with darker skin tones.

The Kramer scale assessments were performed by highly skilled and experienced professionals in this study. In most healthcare settings, particularly in resource-limited areas, assessments may be conducted by less experienced staff, which could impact the real-world performance of VA methods.

Picterus JP also relies on a stable internet connection, a potential limitation in LMICs where connectivity is often inconsistent. Additionally, the study used smartphones with relatively high-resolution cameras, which may not reflect the technology available in more disadvantaged populations. These factors could limit its applicability in these settings. To ensure broader and equitable access, Picterus JP should be compatible with a wider range of smartphones and function reliably with limited or no internet connectivity. Addressing these challenges will help mitigate digital inequalities and enhance the device’s scalability.

Future Research

This study underscores the potential of mHealth technologies like Picterus JP to enhance NNJ screening across diverse populations. Key considerations and future directions include:

1. Accuracy enhancement: Refine the image quality algorithm to better handle higher TSB levels.

2. Diverse validation: Test the device on a wider range of skin types, especially those with higher melanin content.

3. Implementation: Evaluate the feasibility of integrating Picterus JP into health systems, focusing on its impact on reducing NNJ-related morbidity and mortality.

4. Accessibility: Reduce reliance on internet connectivity and increase compatibility with various smartphones for better usability in resource-limited settings.

5. Training: Provide comprehensive training for healthcare providers to improve visual assessment accuracy, particularly in settings where other detection tools are limited.

Conclusions

The study highlights the potential of Picterus JP to support Mexican healthcare workers in identifying newborns at risk of severe NNJ complications. With a significant correlation (ρ = .68) and acceptable accuracy (AUC = 0.875), the device shows potential to be a useful tool for NNJ detection. Appropriate adjustments, such as improving accuracy for high TSB levels and validation in diverse populations, will contribute to enhance Picterus JP’s capability as a screening tool, especially in resource-limited areas. This would contribute to increase early detection and reduce disparities in neonatal health. Continuous investment in research and development, along with collaboration with public health organizations and policy support is essential for wider implementation and scaling up the benefits of mHealth solutions.

Supplemental Material

sj-docx-1-gph-10.1177_30502225251320544 – Supplemental material for Validation of a Mobile Health Device for Neonatal Jaundice Screening: A Cross-Sectional Study in a Resource-Limited Setting in Mexico

Supplemental material, sj-docx-1-gph-10.1177_30502225251320544 for Validation of a Mobile Health Device for Neonatal Jaundice Screening: A Cross-Sectional Study in a Resource-Limited Setting in Mexico by Gabriela Jiménez-Díaz, Lobke M. Gierman, Martina Keitsch, Jesús Elizarrarás-Rivas, Rey Manuel Silva- Méndez, Jennifer J. Infanti and Anna Marcuzzi in Global Pediatric Health

Footnotes

Acknowledgements

The authors wish to express their gratitude to Dr. Yessica Nohemí Rodríguez Sánchez and Dr. Zaida Jatziry Limeta Ortiz from HGZ 1 for their invaluable assistance with data collection. To the IMSS authorities in Mexico for granting us permission to conduct the study in their facilities. Our gratitude goes to Tarik Salem from Picterus AS for for his support with data processing and analysis. Finally, we are deeply grateful to the parents who allowed their newborns to participate in this study.

Correction (July 2025):

This article has been updated with minor textual changes.

ORCID iDs

Gabriela Jiménez-Díaz

Lobke M. Gierman

Martina Keitsch

Jesús Elizarrarás-Rivas

Rey Manuel Silva-Méndez

Jennifer J. Infanti

Anna Marcuzzi

Ethical considerations

Consent to participate

The study objectives were explained to the participants’ parents, who gave their written informed consent prior to the participation. The confidentiality and anonymity of the participants was guaranteed. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

N/A

Author contributions

GJD, JJI, MK, LMG and AM contributed to the conceptualization and design of the study. RMSM, JER, and GJD were involved in the acquisition of the data. GJD, JJI, LMG and AM were responsible for the formal analysis and interpretation of data. GJD wrote the original draft. All authors reviewed and edited the manuscript for important intellectual content and gave final approval. All authors agreed to be accountable for all aspects of the work and resolved any issues related to its accuracy or integrity.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study received funding through an individual PhD grant from the Norwegian University of Science and Technology to the first author. Calibration cards and phones were provided by Picterus AS.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: GJD and LMG are employed and shareholders of Picterus AS which is currently commercializing the medical device Picterus Jaundice Pro.

Data availability

Data underlying the results presented in the study are available under request in (Jiménez-Díaz G, Mendez RMS, Infanti JJ, et al. Data from: validation of a mobile health device for neonatal jaundice screening: a cross-sectional study in a low-resource setting in Mexico. 2024).

Supplemental Material

Supplemental material for this article is available online.

References

Olusanya

Kaplan

Hansen

TWR

. Neonatal hyperbilirubinaemia: a global perspective. Lancet Child Adolesc Health. 2018;2(8):610-620. doi:10.1016/S2352-4642(18)30139-1

Olusanya

Ogunlesi

Slusher

TM.

Why is kernicterus still a major cause of death and disability in low-income and middle-income countries?

Arch Dis Child. 2014;99(12):1117-21. doi:10.1136/archdischild-2013-305506

Bhutani

Wong

Vreman

Stevenson

DK.

Bilirubin production and hour-specific bilirubin levels. J Perinatol. 2015;35(9):735-738. doi:10.1038/jp.2015.32

Bhutani

VK J-HL

. The clinical syndrome of bilirubin-induced neurologic dysfunction. Semin Perinatol. 2011;2011(35):101-103. doi:10.1053/j.semperi.2011.02.003

Bhutani

Zipursky

Blencowe

, et al. Neonatal hyperbilirubinemia and Rhesus disease of the newborn: incidence and impairment estimates for 2010 at regional and global levels. Pediatr Res. 2013;74(Suppl 1):86-100. doi:10.1038/pr.2013.208

Okwundu

Olowoyeye

Uthman

, et al. Transcutaneous bilirubinometry versus total serum bilirubin measurement for newborns. Cochrane Database Syst Rev. 2023;(5):2. doi:10.1002/14651858.CD012660.pub2

Cohen

Wong

Stevenson

DK.

Understanding neonatal jaundice: a perspective on causation. Pediatr Neonatol. 2010;51(3):143-8. doi:10.1016/S1875-9572(10)60027-7

Engle

Jackson

Engle

NG.

Transcutaneous bilirubinometry. Semin Perinatol. 2014;38(7):438-451. doi:10.1053/j.semperi.2014.08.007

Van den Esker-Jonker

den Boer

Pepping

RMC

Bekhof

Transcutaneous bilirubinometry in jaundiced neonates: a randomized controlled trial. Pediatrics. 2016;138(6):e20162414. doi:10.1542/peds.2016-2414

10.

Dräger. Products and Solutions. 2024. https://www.draeger.com/en-us_us/SearchResults?s=Dr%C3%A4ger+Jaundice+Meter+JM-105

11.

UNICEF. Supply Catalogue. 2024. https://supply.unicef.org/s0002644.html

12.

Greco

Arnolda

Boo

N-Y

, et al. Neonatal jaundice in low- and middle-income countries: lessons and future directions from the 2015 don ostrow trieste yellow retreat. Neonatology. 2016;110(3):172-180. doi:10.1159/000445708

13.

Riskin

A TA

Kugelman

Hemo

Bader

Is Visual assessment of jaundice reliable as a screening tool to detect significant neonatal hyperbilirubinemia?

J Pediatr. 2008;152(6):782-787. doi:10.1016/j.jpeds.2007.11.003

14.

Hegde

Rath

Amarasekara

Saraswati

Patole

Rao

Performance of smartphone application to accurately quantify hyperbilirubinemia in neonates: a systematic review with meta-analysis. Eur J Pediatr. 2023;182(9):3957-3971. doi:10.1007/s00431-023-05073-2

15.

Aydin

Hardalac

Ural

Karap

Neonatal jaundice detection system. Journal of Medical Systems. 2016;40(7):166. doi:10.1007/s10916-016-0523-4

16.

Taylor

Stout

de Greef

, et al. Use of a smartphone app to assess neonatal jaundice. Pediatrics. 2017;140(3):e20170312. doi:10.1542/peds.2017-0312

17.

Rong

Luo

, et al. [Evaluation of an automatic image-based screening technique for neonatal hyperbilirubinemia]. Zhonghua Er Ke Za Zhi. 2016;54(8):597-600. doi:10.3760/cma.j.issn.0578-1310.2016.08.008

18.

Ngeow

AJH

Tan

Dong

, et al. Validation of a smartphone-based screening tool (Biliscan) for neonatal jaundice in a multi-ethnic neonatal population. J Paediatr Child Health. 2023;59(2):288-297. doi:10.1111/jpc.16287

19.

Enweronu-Laryea

Leung

Outlaw

, et al. Validating a sclera-based smartphone application for screening jaundiced newborns in Ghana. Pediatrics. 2022;150(1):3600. doi:10.1542/peds.2021-053600

20.

Aune

Vartdal

Bergseng

Randeberg

Darj

Bilirubin estimates from smartphone images of newborn infants’ skin correlated highly to serum bilirubin levels. Acta Paediatr Int J Paediatr. 2020;109(12):2532-2538. doi:10.1111/apa.15287

21.

Jimenez Diaz

. Validation of a new smartphone app to assess neonatal jaundice in a Mexican population. Master thesis. Norwegian University of Science and Technology; 2019. http://hdl.handle.net/11250/2616837

22.

Tusoy

Odland

JØ

Aune

Jimenez-Diaz

A mHealth application as a screening tool for neonatal jaundice in Filipino neonates. Master thesis. Norwegian University of Science and Technology; 2022. https://hdl.handle.net/11250/3006144

23.

Aune

Vartdal

Jimenez Diaz

Gierman

Bergseng

Darj

Iterative Development, Validation, and Certification of a Smartphone System to Assess Neonatal Jaundice: Development and Usability Study. JMIR Pediatr Parent. 2023;6:e40463. doi:10.2196/40463

24.

Worldometer. Accessed July 16, 2024. https://www.worldometers.info/world-population/population-by-country/

25.

Diaz de Leon

Góngora Ortega

eSalud en servicios de salud públicos en México: estudio de caso. Región y Sociedad. 2020;32:e1256. doi:10.22198/rys2020/32/1256

26.

Inegi. 2020 Censo de población y vivienda. Presentacion de Resultados. Oaxaca. 2020. www.inegi.org.mx/contenidos/programas/ccpv/2020/doc/cpv2020_pres_res_oax.pdf

27.

Bujang

Adnan

TH.

Requirements for minimum sample size for sensitivity and specificity analysis. J Clin Diagn Res. 2016;10(10):Ye01-ye06. doi:10.7860/jcdr/2016/18129.8744

28.

Thomas

Soleil et peau [Sun and skin]. Journal of Médecine Esthétique. 1975;2:33-4.

29.

Maya-Enero

Candel-Pau

Garcia-Garcia

Giménez-Arnau

López-Vílchez

Validation of a neonatal skin color scale. Eur J Pediatr. 2020;179(9):1403-1411. doi:10.1007/s00431-020-03623-6

30.

Kramer

LI.

Advancement of dermal icterus in the jaundiced newborn. Am J Dis Child. Sep 1969;118(3):454-458. doi:10.1001/archpedi.1969.02100040456007

31.

Amos

Jacob

Leith

Jaundice in newborn babies under 28 days: NICE guideline 2016 (CG98). Archives of disease in childhood - Education & practice edition. 2017;102(4):207-209. doi:10.1136/archdischild-2016-311556

32.

Kemper

Newman

Slaughter

, et al. Clinical practice guideline revision: management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2022;150(3):58859. doi:10.1542/peds.2022-058859

33.

Diagnóstico y Tratamiento de la Ictericia Neonatal. Guia de Práctica Clínica: evidencias y recomendaciones (CENETEC) (2019).

34.

Keren

Tremont

Luan

Cnaan

Visual assessment of jaundice in term and late preterm infants. Arch Dis Child Fetal Neonatal Ed. 2009;94(5):F317-22. doi:10.1136/adc.2008.150714

35.

Varughese

PM.

Kramer’s scale or transcutaneous bilirubinometry: the ideal choice of a pediatrician? can we trust our eyes?

International Journal of Contemporary Pediatrics. 2019;6(5):1794. doi:10.18203/2349-3291.ijcp20193702

36.

Jurado-SantaCruz F

M-BA

Gutierrez-Vidrio

Ruiz-Rosillo

JM.

Prevalencia del cáncer de piel en tres ciudades de México. Rev Med Inst Mex Seguro Soc. 2011;49(3):253-8.

37.

Swarna

Pasupathy

Chinnasami

Manasa

Ramraj

The smart phone study: assessing the reliability and accuracy of neonatal jaundice measurement using smart phone application. International Journal of Contemporary Pediatrics. 2018;5(2):285. doi:10.18203/2349-3291.ijcp20175928

38.

Ren

Huang

Yang

Gao

The effects on accuracy of image-based estimating neonatal jaundice with a smartphone APP in the different conditions. Pediatric Medicine. 2020;3:AB036. doi:10.21037/pm.2020.AB0

39.

Lingaldinna

Konda

Bapanpally

Alimelu

Singh

Ramaraju

Validity of bilirubin measured by biliscan (smartphone application) in neonatal jaundice – An observational study. Journal of Nepal Paediatric Society. 2021;41(1):93-98. doi:10.3126/jnps.v40i3.29412

40.

Janse

Hoekstra

Jager

, et al. Conducting correlation analysis: important limitations and pitfalls. Clinical Kidney Journal. 2021;14(11):2332-2337. doi:10.1093/ckj/sfab085

41.

Bland

Altman

DG.

Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307-10. doi:doi.org/10.1016/S0140-6736(86)90837-8

42.

Bosschaart

Kok

Newsum

, et al. Limitations and opportunities of transcutaneous bilirubin measurements. Pediatrics. 2012;129(4):689-694. doi:10.1542/peds.2011-2586

43.

De Luca

Jackson

Tridente

Carnielli

Engle

WD.

Transcutaneous bilirubin nomograms: a systematic review of population differences and analysis of bilirubin kinetics. Arch Pediatr Adolesc Med. 2009;163(11):1054-1059. doi:10.1001/archpediatrics.2009.187

44.

Bhutani

Gourley

Adler

Kreamer

Dalin

Johnson

LH.

Noninvasive measurement of total serum bilirubin in a multiracial predischarge newborn population to assess the risk of severe hyperbilirubinemia. Pediatrics. 2000;106(2):e17-e17. doi:10.1542/peds.106.2.e17

45.

Varughese

Krishnan

Does color really matter? Reliability of transcutaneous bilirubinometry in different skin-colored babies. Indian Journal of Paedia-tric Dermatology. 2018;19(4):315-320. doi:10.4103/ijpd.IJPD_3_18

46.

Thomas

Warner

Jones

GRD

, et al. Total bilirubin assay differences may cause inconsistent treatment decisions in neonatal hyperbilirubinaemia. Clin Chem Lab Med. 2022;60(11):1736-1744. doi:10.1515/cclm-2022-0749

47.

Hansen

TWR

. Narrative review of the epidemiology of neonatal jaundice. Pediatric Medicine. 2021;4(18):1-14. doi:10.21037

48.

Estrategia Digital Nacional . 2013:44. https://www.gob.mx/cms/uploads/attachment/file/17083/Estrategia_Digital_Nacional.pdf

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB