Understanding the risk factors for adverse events during exchange transfusion in neonatal hyperbilirubinemia using explainable artificial intelligence

doi:10.21203/rs.3.rs-1537494/v1

Download PDF

Research Article

Understanding the risk factors for adverse events during exchange transfusion in neonatal hyperbilirubinemia using explainable artificial intelligence

https://doi.org/10.21203/rs.3.rs-1537494/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Objective

To understand the risk factors associated with adverse events during exchange transfusion (ET) in severe neonatal hyperbilirubinemia.

Study Design

We conducted a retrospective study of infants with hyperbilirubinemia who underwent ET within 30 days of birth from 2015 to 2020 in a children’s hospital. Both traditional statistical analysis and state-of-the-art explainable artificial intelligence (XAI) were used to identify the risk factors.

Results

A total of 188 ET cases were included; 7 major adverse events, including hyperglycemia (86.2%), top-up transfusion after ET (50.5%), hypocalcemia (42.6%), hyponatremia (42.6%), thrombocytopenia (38.3%), metabolic acidosis (25.5%), and hypokalemia (25.5%), and their risk factors were identified. Some novel and interesting findings were identified by XAI.

Conclusions

XAI not only achieved better performance in predicting adverse events during ET but also helped clinicians to more deeply understand nonlinear relationships and generate actionable knowledge for practice.

Jaundice is a common condition in neonates. Approximately 60% and 80% in term and preterm infants have clinical jaundice in the first week after birth, but only very small part of them (0.02% and 0.16% in term or preterm) develop severe hyperbilirubinemia[1]. Severe neonatal hyperbilirubinemia can cause neurological disability such as encephalopathy or mortality if not effectively managed[2]. Phototherapy and exchange transfusion (ET) are the primary treatment modalities to prevent bilirubin encephalopathy[3, 4]. Although ET has become a rare event in most developed countries, it remains a frequent emergency rescue procedure for severe neonatal hyperbilirubinemia especially in many developing countries[5, 6]. ET is effective and considered to be a safe procedure; however, it is not without risks, and the mortality rates range from 0.5–3.3% in different studies[7]. Therefore, the current recommendations for performing ET are based on a balance between the risks of encephalopathy and adverse events related to the procedure. Common adverse events during ET include hyperglycemia, thrombocytopenia, hypocalcemia, hypokalemia or hyperkalemia, and metabolic acidosis, can be monitored and corrected in a timely manner[8]. Although some risk factors for these adverse events have been reported in several studies[7–10], only risk factors with a linear relationship have been identified for these adverse events during ET using traditional data analysis. The purpose of this study was to evaluate these adverse events during ET in neonatal hyperbilirubinemia and to identify the potential risk factors for these complications based on state-of-the-art explainable artificial intelligence (XAI) technology.

Although linear models have historically been popular because they are interpretable, modern complex machine learning models often achieve higher predictive accuracy because they capture interactions among variables, in addition to nonlinear relationships[11, 12]. In addition to the superior performance of modern machine learning models, some explainable artificial intelligence techniques, such as SHAP (SHapley Additive exPlanations), can better demonstrate nonlinear relationships (e.g., U-shaped relationships)[13, 14], and new relationships discovered by the model are even more valuable than the application of the model itself. In particular, the discovery of new relationships can help medical professionals to control some avoidable risks or prepare in advance for specific unavoidable risks.

The medical records of neonates who received exchange transfusions to treat severe hyperbilirubinemia in neonatal units at the Children’s Hospital, Zhejiang University School of Medicine for a period of six years (from January 2015 through December 2020) were reviewed retrospectively. The indications for ET and the method of ET followed the relevant Chinese clinical guidelines[15], based on which the double volume exchange method (150–160 ml/kg) was completed for approximately 90–120 minutes. Blood gas, blood glucose, electrolytes, blood calcium, and blood cell counts were monitored during ET.

Patient characteristics, such as sex, gestational age, delivery mode, Apgar score at birth, weight at birth, weight at admission, age at admission, parents’ and baby’s blood group, mode of feeding before ET, the artery and vein used for ET, and relevant laboratory tests, such as direct bilirubin (DBIL), indirect bilirubin (IBIL) and total bilirubin (TBIL), serum calcium, glucose, sodium, potassium, white blood cells, hemoglobin, pH, and HCO₃, were collected at different time points before, during and after ET.

The adverse events during ET were defined by the following criteria. Hyperglycemia occurred when serum glucose was > 7.2 mmol/L, metabolic acidosis if HCO₃ was < 18 mmol/L, hyperkalemia if serum potassium was ≥ 5.5 mmol/L, hypokalemia when serum potassium was < 3.0 mmol/L, hypocalcemia if serum calcium was < 0.9 mmol/L, thrombocytopenia if platelet count < 100×10⁹/L, hyponatremia if serum sodium was < 135 mmol/L, cyanosis if SpO₂ < 90%, and top-up transfusion if hemoglobin reduction met clinical indications for transfusion.

The neonates were categorized according to their status with/without specific adverse events during ET. Continuous variables (such as age and weight) of the patients were reported as the mean ± SD and were compared using the Mann-Whitney U test. Categorical variables (such as sex) were reported as counts (percentages) and compared using the chi-square test. A p value < 0.05 was considered statistically significant.

Several widely used machine learning methods, such as random forest, XGBoost, logistic regression, Gaussian naïve Bayes, and k-neighbors, were used to train a machine learning model using 70% of the data and test them on the standby 30% of the data, which were split randomly. These five machine learning models were from the scikit-learn Python package. In this study, the best performing model was enhanced with an interpretation method called SHAP[13], which is a game-theoretic approach for explaining the output of any machine learning model by computing each feature for the prediction. It calculates exact SHAP values for each feature. In this study, higher SHAP values imply large contributions to adverse event risk. This explainable machine learning model helps clinicians to understand the risk factors for a single prediction, for a single variable and for the entire dataset. These explanations have the potential to generate human actionable knowledge to improve clinical outcomes.

There were 188 exchange transfusions performed in 185 neonates in this study, as shown in Table 1. Among them, 112 (59.6%) were male. Overall, 34 (18.1%) infants were preterm, and the mean gestational age was 37.93 ± 1.63 weeks old. The mean age at admission was 6.42 ± 3.97 days old, and exchange transfusion was performed 6.63 ± 3.40 hours after admission. Among 188 cases, ABO incompatibility was found in 65 infants (34.6%), RH incompatibility in 21 (11.2%), G6PD deficiency in 30 (16.0%), and MN incompatibility in 1 (0.5%). Bilirubin encephalopathy was diagnosed in 63 (33.5%) cases, sepsis was diagnosed in 22 (11.7%), anemia was diagnosed in 15 (8%), and NEC (necrotizing enterocolitis) and purulent meningitis were diagnosed in 2 each (1.1%). Among all cases, 185 (98.4%) experienced different adverse events, and the most common adverse events were hyperglycemia (86.2%), followed by anemia requiring top-up transfusion after ET (50.5%), hypocalcemia (42.6%), hyponatremia (42.6%), thrombocytopenia (38.3%), acidosis (25.5%), hypokalemia (25.5%), hyperkalemia (3.2%), convulsions (2.7%) and cyanosis (2.7%). Considering the sample size, this study focused on only the 7 most common adverse events.

Table 1

Baseline demographic characteristics
Characteristic	Value
Total ET count	188
Male sex	112(59.6%)
Gestational age (weeks)	37.93 ± 1.63
Premature	34 (18.1%)
Delivery by C-section	66(35.1%)
Apgar score	9.88 ± 0.48
Birth weight (g)	3201.30 ± 429.75
Weight at admission (g)	3036.97 ± 430.53
Age at admission (days)	6.42 ± 3.97
Length of stay	9.31 ± 3.93
Etiology
Unidentified	96(51.1%)
ABO incompatibility	65(34.6%)
G6PD deficiency	30(16.0%)
RH incompatibility	21 (11.2%)
MN incompatibility	1 (0.5%)
Other diagnosis
Bilirubin encephalopathy	63 (33.5%)
Sepsis	22(11.7%)
Anemia	15 (8%)
Necrotizing enterocolitis	2(1.1%)
Purulent meningitis	2(1.1%)
Exchange transfusion (ET)
ET volume (ml)	495.27 ± 78.59
ET time (minutes)	96.77 ± 18.82
TSB before ET	388.70 ± 107.66
TSB during ET	253.41 ± 80.29
TSB after ET	198.75 ± 62.13
TSB 1 day after ET	220.60 ± 58.95
Bilirubin exchange rate	0.48 ± 0.10
Adverse events during ET
Hyperglycemia	162(86.2%)
Requiring top-up transfusion after ET	95(50.5%)
Hypocalcemia	80(42.6%)
Hyponatremia	80(42.6%)
Thrombocytopenia	72(38.3%)
Metabolic acidosis	48(25.5%)
Hypokalemia	48(25.5%)
Hyperkalemia	6(3.2%)
Convulsion	5(2.7%)
Cyanosis	5(2.7%)

The variables with significant differences between patients with and without 7 common adverse events during ET are shown in Table 2. Please note that directly related variables regarding the adverse events, such as blood glucose for hyperglycemia and serum calcium for hypocalcemia, are not shown in this table. Younger age is a common risk factor for these adverse events, and there are significant differences in infants with/without thrombocytopenia, hypokalemia, top-up transfusion, and hypocalcemia. However, the term and preterm infants did not show significant differences in these adverse events. Female neonates are more likely to experience hypokalemia. Breastfeeding can reduce the risk of hypokalemia and hypocalcemia. Not starting feeding is a risk factor for metabolic acidosis and hypocalcemia. Formula feeding increases the risk of hypocalcemia. There was also no significant difference in the mode of delivery or Apgar score for adverse events during ET. Different etiologies of hyperbilirubinemia have different risks for different adverse reactions. ABO incompatibility is associated with a higher risk of metabolic acidosis but a lower risk of hyperglycemia and hyponatremia. RH incompatibility is associated with a higher risk of hypocalcemia. G6PD deficiency reduces the risk of hypocalcemia and hypokalemia. Etiology-identified neonates more likely to experience top-up transfusion after ET. A short ET time contributes to hyperglycemia and hypocalcemia. The ET speed showed a significant difference between infants with and without hypocalcemia. ET via the femoral artery will increase the risk of hypocalcemia. ET via the axillary artery, femoral vein or popliteal vein could decrease the risk of hyperglycemia, hyponatremia and hypokalemia, respectively. Infants with higher white blood cell counts before ET have a higher risk of hypocalcemia and metabolic acidosis. A relatively lower TBIL level is significantly associated with many adverse events, such as thrombocytopenia, hypokalemia, top-up transfusion, and hypocalcemia. The relationships among different adverse events are shown in the alluvial diagram (Fig. 1h).

Table 2

Comparing variables with and without adverse events during exchange transfusion
adverse event	Variables	NO	Yes	p value
hyperglycemia 162(86.2%)	ABO incompatibility	15(57.7%)	50(30.9%)	0.014
	ET time	104.88 ± 20.81	95.47 ± 18.22	0.017
	ET artery axillary	5(19.2%)	5(3.1%)	0.003
	Serum potassium	3.87 ± 0.39	3.65 ± 0.53	0.046
	Cerebral hemorrhage	1(3.8%)	42(25.9%)	0.025
top-up transfusion after ET 95(50.5%)	Admission age	7.00 ± 4.04	5.85 ± 3.83	0.046
	Etiology unidentified	57(61.3%)	39(41.1%)	0.009
	IBIL at admission	431.34 ± 99.24	394.97 ± 113.56	0.021
	TBIL_bga at admission*	528.25 ± 111.54	465.65 ± 122.34	< 0.001
	TBIL_bga before ET	407.01 ± 107.13	370.78 ± 105.68	0.021
	Hemoglobin	142.83 ± 33.67	121.18 ± 29.40	< 0.001
Hypocalcemia 80(42.6%)	Admission age	7.51 ± 3.64	4.94 ± 3.93	< 0.001
	Breast feeding	76(70.4%)	37(46.2%)	0.001
	Formula feeding	9(8.3%)	21(26.2%)	0.002
	No feeding	0(0.0%)	7(8.8%)	0.006
	RH incompatibility	3(2.8%)	18(22.5%)	< 0.001
	G6PD deficiency	25(23.1%)	5(6.2%)	0.003
	Etiology unidentified	64(59.3%)	32(40.0%)	0.014
	ET time	99.31 ± 19.56	93.35 ± 17.32	0.032
	ET speed	5.09 ± 0.93	5.44 ± 1.02	0.014
	ET artery femoral	21(19.4%)	29(36.2)	0.016
	TBIL_bga at admission	528.69 ± 95.60	453.31 ± 137.64	< 0.001
	TBIL_bga before ET	409.50 ± 103.33	360.62 ± 107.63	0.002
	Serum bicarbonate	21.83 ± 2.60	20.64 ± 2.71	0.003
	White cell count	12.97 ± 4.77	17.01 ± 10.59	0.001
hyponatremia 80(42.6%)	ABO incompatibility	45(41.7%)	20(25.0%)	0.026
	ET vein femoral	21(19.4%)	5(6.2%)	0.017
	Bilirubin exchange rate	0.46 ± 0.11	0.51 ± 0.08	0.004
thrombocytopenia 72(38.3%)	Admission age	6.93 ± 4.14	5.58 ± 3.55	0.023
	TBIL at admission	453.56 ± 119.07	403.52 ± 102.60	0.004
	TBIL_bga Before ET	404.76 ± 109.88	362.83 ± 99.35	0.009
	Hemoglobin	137.72 ± 32.73	122.49 ± 32.29	0.002
Metabolic Acidosis 48(25.5%)	Gravidity	G2(34.8%);G3(15.6%)	G2(13.3%);G3(31.1%)	0.021
	Not feeding	2 (1.4%)	5 (10.4%)	0.017
	ABO incompatibility	40(28.6%)	25(52.1%)	0.005
	Serum calcium	1.17 ± 0.12	1.13 ± 0.11	0.025
	White cell count	13.98 ± 6.92	16.75 ± 10.43	0.039
hypokalemia 48(25.5%)	Sex (female)	50(35.7%)	26(54.2%)	0.038
	Admission age	7.00 ± 3.98	4.73 ± 3.42	0.001
	Breast feeding	91(65.0%)	22(45.8%)	0.030
	G6PD deficiency	28(20.0%)	2(4.2%)	0.018
	ET vein popliteal	20(14.5%)	0(0.0%)	0.010
	TBIL_bga before ET	398.06 ± 105.39	361.42 ± 110.65	0.042
	Serum calcium	1.18 ± 0.11	1.11 ± 0.14	0.001
*varabile with subscript bga means the variable were measured by blood gas analyzer to distinguished it from biochemistry value

Among the five machine learning models, the XGBoost model achieved the best performance in the prediction tasks of 7 adverse events (detailed information in Supplemental Tables S1 and S2). Using the SHAP, the top 10 features contributing to the 7 adverse events are shown in Fig. 1a-g. Each dot in each feature corresponds to an individual case in the dataset. The position of a dot on the horizontal axis indicates the impact of the feature (SHAP value) on model prediction, and the color of a dot reflects the feature value of the case. Not surprisingly, all indicators directly related to adverse events (e.g., blood glucose before ET and hyperglycemia) played the most important roles in the prediction models for each adverse event. In addition to variables with statistically significant differences, a number of nonlinear relationships were identified in the interpretable AI model that will help clinical staff to understand these risks in greater depth. The detailed relationships of the top 10 features identified by XAI are shown in Supplemental Figures S1-S7.

For hyperglycemia, both the statistical analysis and XAI identified that a high blood glucose level before ET, ABO incompatibility, ET time, and serum potassium are important risk factors. XAI also shows that platelet count and ET volume associated with hyperglycemia. In Fig. 2a-b, these two variables both show a nonlinear relationship. Both large and small ET volumes increase the risk of hyperglycemia. Only a small range of approximately 500 ml will decrease the risk. Unexpectedly lower platelet counts, especially with lower blood glucose levels, result in a higher hyperglycemia risk. However, higher platelet counts with lower blood glucose levels can reduce the risk.

For top-up transfusion after ET, the statistical analysis only demonstrated that lower IBIL and TBIL contribute to this adverse event, but XAI identified that higher DBIL in infants and lower TBIL contribute to this adverse event.

Although there were significant differences between different feeding modes with/without hypercalcemia, XAI did not include them as significant factors in the prediction of hypocalcemia. Traditional statistical analysis focused on the short ET time and higher ET speed, while XAI showed that a bilirubin exchange rate of ET > 0.5 was a more important risk factor for hypocalcemia, as shown in Fig. 3a. The risk of hypocalcemia decreases initially with increasing ET speed but increases significantly when the ET speed exceeds 6.3 ml/min, as shown in Fig. 3b. Both XAI and traditional analysis show that an elevated white cell count is a risk factor for hypocalcemia. XAI also identified the relationship between pH and the risk of hypercalcemia.

Except for the bilirubin exchange rate among the 3 risks identified by statistical analysis for hyponatremia, XAI showed more complex relationships between variables and hyponatremia, such as a cliff-like pattern change with a platelet count of approximately 300 ×10⁹/L. HCO₃ also showed a reverse U-shaped relationship with hyponatremia.

Except for the lower hemoglobin identified by statistical analysis, higher serum bicarbonate, higher serum potassium, higher white cell count, and lower serum calcium all contribute to thrombocytopenia based on XAI.

An unexpected relationship between metabolic acidosis and the waiting time until ET after admission was identified, as shown in Fig. 4a. It seems that performing ET 5 hours after admission will help to control the occurrence of acidosis. The third gravidity was related to a higher risk of acidosis during ET (Fig. 4b). From the original data, there were 40% third gravidity infants vs. 11% second gravidity infants with metabolic acidosis in this cohort.

Although popliteal vein infusion was identified as a risk factor for hypokalemia in statistical analysis, the XAI considered the radial artery to be a more important feature. An interesting finding is that the diagnosis of bilirubin encephalopathy reduced the risk of hypokalemia. The G6PD deficiency, which showed a significant difference in statistical analysis, was not among the top 10 features identified by XAI.

This study demonstrated that explainable artificial intelligence not only achieves better performance but can also help clinicians to more deeply understand the nonlinear relationships among various clinical indicators and outcomes.

However, some factors did not show significant differences in traditional statistical analysis, such as the XAI-identified elevation of direct bilirubin increasing the risk of top-up transfusion after ET. This result was also supported by a basic study that showed that direct bilirubin triggers anemia[16]. Since direct bilirubin and total bilirubin show opposite effects on the risk of top-up transfusion, the ratio of direct bilirubin to total bilirubin could be used as a more significant indicator of the associated risk.

A clear threshold rather than a traditional correlation is more conducive to gaining actionable clinical knowledge to effectively control risk. XAI, which can present a clear threshold, as shown in Fig. 3, offers another advantage of using it to analyze clinical data.

XAI could identify some novel relationships that could be missed by traditional statistical analysis due to nonlinear relationships or simple distribution imbalance issues. Such novel relationships should inspire researchers to further study them.

In conclusion, we used traditional statistical analysis and XAI to identify the risk factors for 7 major adverse events during exchange transfusion in neonatal hyperbilirubinemia. The XAI model achieved better performance in predicting adverse events and provided more useful and actional knowledge for clinicians.

ET Exchange Transfusion

XAI Explainable Artificial Intelligence

SHAP SHapley Additive exPlanations

TBIL Total Bilirubin

IBIL Indirect Bilirubin

DBIL Direct Bilirubin

G6PD Glucose-6-Phosphate Dehydrogenase

Ethics approval and consent to participate

This study was approved by the Institutional Review Board/Ethics Committee (2022-IRB-069) of the Children’s Hospital, Zhejiang University School of Medicine, and the study was performed in accordance with the Declaration of Helsinki. Written informed consent was waived by the Institutional Review Board/Ethics Committee since the utilization of anonymized retrospective data does not require patient consent under local legislation.

Consent for publication

Not applicable

Availability of data and materials

All data generated or analyzed during this study are included in this article and its supplementary material files. Further enquiries can be directed to the corresponding author.

Competing interests

The authors declare that they have no conflicts of interest.

Funding

This work was supported by the National Natural Science Foundation of China (81871456).

Authors’ contributions

SZ and HL designed the experiments and wrote the manuscript. SZ, LZ, YF, JZ, QS and HL collected the original data. SZ, YF, QS and HL performed the research. JZ, QS and HL reviewed the manuscript.

Acknowledgements

We acknowledge the support of the Children’s Hospital of Zhejiang University School of Medicine (Zhejiang, China) for supplying the anonymized clinical data.

Watchko JF, Maisels MJ. Management of severe hyperbilirubinemia in the cholestatic neonate: a review and an approach. J Perinatol 2022. 2022;:1–7. doi:10.1038/s41372-022-01330-8.
Brites D, Fernandes A, Falcão AS, Gordo AC, Silva RFM, Brito MA. Biological risks for neurological abnormalities associated with hyperbilirubinemia. J Perinatol. 2009;29:S8–13.
Murki S, Kumar P. Blood Exchange Transfusion for Infants with Severe Neonatal Hyperbilirubinemia. Semin Perinatol. 2011;35:175–84.
Smitherman H, Stark AR, Bhutan VK. Early recognition of neonatal hyperbilirubinemia and its emergent management. Semin Fetal Neonatal Med. 2006;11:214–24.
Owa JA, Ogunlesi TA. Why we are still doing so many exchange blood transfusion for neonatal jaundice in Nigeria. World J Pediatr. 2009;5:51–5.
Zahed Pasha Y, Alizadeh-Tabari S, Zahed Pasha E, Zamani M. Etiology and therapeutic management of neonatal jaundice in Iran: a systematic review and meta-analysis. World J Pediatr. 2020;16:480–93.
Sabzehei MK, Basiri B, Shokouhi M, Torabian S. Complications of Exchange Transfusion in Hospitalized Neonates in Two Neonatal Centers in Hamadan, A Five-Year Experience. J Compr Pediatr. 2015;6.
Patra K, Storfer-Isser A, Siner B, Moore J, Hack M. Adverse events associated with neonatal exchange transfusion in the 1990s. J Pediatr. 2004;144:626–31.
Chacham S, Kumar J, Dutta S, Kumar P. Adverse events following blood exchange transfusion for neonatal hyperbilirubinemia: A prospective study. J Clin Neonatol. 2019;8:79.
Behjati S, Sagheb S, Aryasepehr S, Yaghmai B. Adverse events associated with neonatal exchange transfusion for hyperbilirubinemia. Indian J Pediatr. 2009;76:83–5.
Coudray N, Ocampo PS, Sakellaropoulos T, Narula N, Snuderl M, Fenyö D, et al. Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning. Nat Med. 2018;24:1559–67.
Tomašev N, Harris N, Baur S, Mottram A, Glorot X, Rae JW, et al. Use of deep learning to develop continuous-risk models for adverse event prediction from electronic health records. Nat Protoc. 2021;16:2765–87.
Zhang K, Zhang Y, Wang M. A Unified Approach to Interpreting Model Predictions Scott. Nips. 2012;16:426–30.
Lundberg SM, Erion G, Chen H, DeGrave A, Prutkin JM, Nair B, et al. From local explanations to global understanding with explainable AI for trees. Nat Mach Intell. 2020;2:56–67.
Neonatology Group of Pediatrics Branch of Chinese Medical Association. Expert consensus on the diagnosis and treatment of neonatal hyperbilirubinemia. Chinese J Pediatr. 2014;52:745–8.
Lang E, Gatidis S, Freise NF, Bock H, Kubitz R, Lauermann C, et al. Conjugated bilirubin triggers anemia by inducing erythrocyte death. Hepatology. 2015;61:275–84.

No competing interests reported.

Supplementalmaterial.docx

Download PDF

Editorial decision: Major revision
05 Jul, 2022
Reviews received at journal
27 Jun, 2022
Reviewers agreed at journal
09 Jun, 2022
Reviews received at journal
03 Jun, 2022
Reviewers agreed at journal
21 May, 2022
Reviews received at journal
08 May, 2022
Reviewers agreed at journal
21 Apr, 2022
Reviewers agreed at journal
20 Apr, 2022
Reviewers invited by journal
11 Apr, 2022
Editor assigned by journal
11 Apr, 2022
Editor invited by journal
11 Apr, 2022
Submission checks completed at journal
11 Apr, 2022
First submitted to journal
08 Apr, 2022

You are reading this latest preprint version

Understanding the risk factors for adverse events during exchange transfusion in neonatal hyperbilirubinemia using explainable artificial intelligence

Status:

Version 1

Abstract

Figures

Introduction

Subjects And Methods

Results

Discussion

Abbreviations

Declarations

Ethics approval and consent to participate

Consent for publication

Availability of data and materials

Competing interests

Funding

Authors’ contributions

Acknowledgements

References

Additional Declarations

Supplementary Files

Status:

Version 1