Utility Of Support Vector Machine And Decision Tree To Identify The Prognosis Of Metformin Poisoning In The United States: Analysis Of National Poisoning Data System

doi:10.21203/rs.3.rs-1330787/v1

Download PDF

Research Article

Utility Of Support Vector Machine And Decision Tree To Identify The Prognosis Of Metformin Poisoning In The United States: Analysis Of National Poisoning Data System

https://doi.org/10.21203/rs.3.rs-1330787/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: With diabetes incidence growing globally, and metformin still being the first-line for its treatment, metformin’s toxicity and overdose have been increasing. Hence, its mortality rate is increasing. For the first time, we aimed to study the efficacy of machine learning algorithms in predicting the outcome of metformin poisoning using two well-known classifications methods including support vector machine (SVM) and decision tree (DT).

Methods: This is a retrospective cohort study of National Poison Data System data which is the largest data repository of poisoning cases in the United States. The SVM and DT algorithms developed using training and test datasets.

Results: The overall accuracy of the SVM and DT model were 85% and 89%, respectively. Our model showed that acidosis, hypoglycemia, electrolyte abnormality, hypotension, elevated anion gap, elevated creatinine, tachycardia, and renal failure are the most important determinants in terms of outcome prediction of metformin poisoning. Furthermore, we found that the accuracy of the SVM was higher than the DT model, so it could perform more efficiently to predict the outcomes.

Conclusions: In order to predict the prognosis of metformin poisoning, machine learning algorithms are very effective, and may help clinicians in the process of management and follow up of the metformin poisoning cases.

metformin

National Poison Data System

support vector machine

decision tree

Diabetes has become a public health concern that gives rise to serious macrovascular and microvascular complications [1]. In 2015, there were approximately 415 million diabetes patients worldwide; this figure is expected to grow to over 600 million by 2040 [2]. Metformin is still the first-line therapy for diabetes among all medicines currently available [3]. However, even though metformin is a safe drug, poisoning can have a fatality rate of 30–50%, climbing to as much as 80% if done intentionally [4]. Due to the lack of an antidote and a conservative approach to managing metformin poisoning, early prognosis prediction based on initial presentation, might be critical in diminishing the death rate.

Medicine has paid a lot of attention lately to new classification methods [5]. The fascinating part about machine learning techniques is that they can be taught how to assess medical risk predictions and simulate complicated clinical scenarios [6]. Thus, machine learning could potentially boost precision medication delivery more than regression analysis currently does [7]. Surprisingly, massive amounts of data heterogeneity can be combined using machine learning to uncover causal relationships between diseases and classify risk variables [8]. The decision tree (DT) examines the input data and shows the outcomes of the predicted relationships that have been identified as a consequence of the implementation of appropriate rules [9]. When dealing with huge amounts of variables and limited sample size in medical research, the support vector machine (SVM) is another promising classification model [10]. SVM is a breakthrough tool that outputs findings while establishing a hyperplane, culminating in classification. Essentially, the underlying concept of SVM is to establish decision boundaries wherein characteristics are assessed concerning a hyperplane to demonstrate their significance. The attributes for which the hyperplane is drawn and the distance between the hyperplane and the attributes are called support vectors and margins, respectively. The optimal hyperplane is the one with the largest margin of support vectors.

The application of DT and SVM have been widely investigated among patients with drug overdose in the recent times [11–13]. However, despite the serious outcomes, there has been no explicit application of machine learning techniques to study metformin overdose and to provide a realistic strategy for determining the prognosis based on clinical and laboratory findings. Besides, although many of the important features of metformin poisoning are well known, the use of machine learning algorithms to determine metformin poisoning is not studied yet. This is the first study that used machine learning algorithms in outcome prediction of metformin poisoning. We aimed to apply and test the decision tree and support vector machine models on a broad scale data of patients with metformin overdose, taken from the National Poison Data System (NPDS), to evaluate the different categories of outcomes.

Study Population and Eligibility Criteria

The data of this study was obtained from NPDS, which is the only real-time data repository of poisoning in the United States. American Association of Poison Control Centers (AAPCC) which maintains the NPDS, represents the 55 Poison Control Centers (PCCs). NPDS includes exposures to more than 400,000 substances that have been continuously reported by PCCs [14]. Notably, more than two million human exposure was reported to NPDS in 2019 [14]. Even though NPDS data does not contain all of the substance exposure in the country, every exposure reported to NPDS does not necessarily signify poisoning or toxicity. All metformin exposure cases reported to the NPDS between January 1, 2012, and December 31, 2017, were included in this study. However, we excluded those cases with missing data as well as duplicate ones. Institutional review board approval was not required for this study based on the Colorado Multiple Institutional Review Board on Human Subjects Protection standards. All methods were carried out in accordance with relevant guidelines and regulations.

Definition Of Terms

To develop our classification model, we defined some important features based of the NPDS guidelines as follows:

Hypertension: Diastolic blood pressure greater than 90 mm Hg or systolic blood pressure greater than 140 mm Hg

Elevated anion gap: Result of the following equation more than 12 mEq/L: [Na+ - (Cl- + HCO3-)]

Elevated creatinine: Creatinine level more than 1.5 mg/dL or 133 µmol/L

Tachycardia: Heart rate more than 100 beats per minute

Renal failure: Acute and chronic renal failure that leads to clinically substantial loss of renal function and azotemia

Electrolyte abnormality: Imbalance level of sodium, potassium, bicarbonate, chloride, calcium, magnesium, and phosphate

Hypoglycemia: Glucose levels of less than 70 mg/dL or 3.9 mmol/L

Acidosis: Bicarbonate level less than 20 mEq/L, pH less than 7.35, or elevated levels of lactic acid

Development Of Classification Model

First, the dataset was randomly divided into training (70%) and testing (30%) datasets. The prediction model was developed by utilizing a train set and then incorporating the various variables, including demographic data (age, sex), the purpose of exposure (suicidal, unintentional, etc.), chronicity, clinical features, etc. Next, the test set was utilized to evaluate the model performance to see how well it fits the training set. Every decision tree comprises some nodes, including root and leaf nodes and branches. The root node denotes the most important feature, whereas the leaf node depicts a decision by applying some IF-THEN rules. For example, when moving down the decision tree's path, the right and left directions indicate false and true, respectively. The evaluation of the decision tree model was performed through F-1 score, specificity, recall, precision, accuracy, and confusion matrix. All of the analyses were done in Python using the Sklearn library.

Prognosis Prediction Based on Decision Tree

A total of 2878 cases with metformin exposure were included in this study. The decision tree model and 15 rules driven from it are shown in Fig. 1 and Table 1, respectively. The most important feature of our model is shown in Fig. 2 as well. Our decision model comprises of ten levels, fifteen leaf nodes, and a total of 29 nodes. Acidosis was the most prominent determinate of the metformin exposure prognosis, followed by hypoglycemia and electrolyte abnormality. Evaluation of the training and test datasets showed that although the significant levels of recall (training set: 98.4%; test set: 98.2%) and F1-score (training set: 91.5%; test set: 91.2%) attributed to minor outcomes, the greatest specificity (training set: 98.5%; test set: 98.4%) and precision (training set: 87.9%; test set: 87.4%) belonged to major and moderate outcomes, respectively (Tables 2 and 3). This evaluation also demonstrated the accuracy of 85% and 84% for training and test datasets, respectively. The confusion matrix showed that our model successfully identifies 1201 cases of minor outcomes, 449 cases of moderate outcomes, and 74 cases of major outcomes in the training set and 518 cases of minor outcomes, 188 cases of moderate outcomes, and 28 cases of major outcomes in test datasets (Tables 4 and 5).

Table 1

The Rules Driven from The Decision Tree Model
If acidosis and renal failure occur, patients will develop a major outcome (100%)	1
If acidosis and hypotension occur while renal failure does not occur, patients are more likely to develop major outcomes (64.6%)	2
If acidosis and hypoglycemia are present, while renal failure and hypotension do not occur, patients are more likely to develop major outcomes (50%)	3
If acidosis occurs while renal failure, hypotension and hypoglycemia do not occur, patients are more likely to develop moderate outcomes (86%)	4
If a patient experiences hypoglycemia without acidosis, moderate outcomes are more likely (92.9%)	5
If a patient with an age of more than 59.5 experiences electrolyte abnormalities without acidosis, hypoglycemia, the development of major outcomes is more likely (66.7%)	6
If a patient with an age of less than 59.5 experiences electrolyte abnormalities without acidosis, hypoglycemia, the development of a moderate outcome is very likely (96.9%)	7
If tachycardia and hypertension are present, while acidosis, hypoglycemia and electrolyte abnormalities are not present, patients develop moderate outcomes (100%)	8
If tachycardia is present without acidosis, hypoglycemia, electrolyte abnormalities and hypertension, patients are likely to develop a minor result (50%)	9
If increased creatinine is present, while acidosis, hypoglycemia, electrolyte abnormalities and tachycardia are not present, patients are more likely to develop moderate outcomes (90%)	10
If an elevated anion gap is present, while acidosis, hypoglycemia, electrolyte abnormalities, tachycardia and elevated creatinine are not present, patients develop moderate outcomes (100%)	11
If a patient has an unintentional exposure without hypoglycemia, electrolyte abnormalities, tachycardia, elevated creatinine and elevated anion gap, the patient is more likely to develop minor outcomes (94.6%)	12
If acidosis does not occur, hypoglycemia does not occur, electrolyte abnormality does not exist, tachycardia does not exist, elevated creatinine does not exist, elevated anion gap does not exist, the reason for exposure is not unintentional, other miscellaneous are present, THEN patients are more likely to develop minor outcomes (71%)	13
If hypotension occurs, while acidosis, hypoglycemia, electrolyte abnormalities, tachycardia, elevated creatinine, elevated anion gap do not occur, the reason for exposure is unintentional, other miscellaneous are not present, THEN patients are more likely to develop moderate outcomes (75%)	14
If the cause of exposure is unintentional, and acidosis, hypoglycemia, electrolyte abnormalities, tachycardia, elevated creatinine, elevated anion gap and hypotension do not occur, other miscellaneous are not present, patients are more likely to develop minor outcomes (86.5%)	15

Table 2

Decision Tree Characteristics for The Training Set
Labels	Major Effect	Minor Effect	Moderate Effect	Average	Weighted Average
Specificity	0.985169	0.744332	0.954681	0.894728	0.829168
Precision	0.725490	0.855413	0.879921	0.820275	0.855414
Recall	0.587302	0.984426	0.669162	0.746963	0.855015
F1_score	0.649123	0.915396	0.760204	0.774908	0.847264
Accuracy	NaN	NaN	NaN	0.855015	0.855015

Table 3

Decision Tree Characteristics for Test Set
Labels	Major Effect	Minor Effect	Moderate Effect	Average	Weighted Average
Specificity	0.984029	0.732938	0.953206	0.890058	0.820636
Precision	0.682927	0.851974	0.874419	0.803106	0.849647
Recall	0.560000	0.982922	0.655052	0.732658	0.849537
F1_score	0.615385	0.912775	0.749004	0.759055	0.841164
Accuracy	NaN	NaN	NaN	0.849537	0.849537

Table 4

Confusion Matrix in DT Model for The Training Set
Prediction	Major Effect	Minor Effect	Moderate Effect
True
Major Effect	74	9	43
Minor Effect	1	1201	18
Moderate Effect	27	194	447
0.85

Table 5

Confusion Matrix in DT Model for Test Set
Prediction	Major Effect	Minor Effect	Moderate Effect
True
Major Effect	28	4	18
Minor Effect	0	518	9
Moderate Effect	13	86	188
0.84

Prognosis Prediction Based On Support Vector Machine

Evaluation of the training dataset of SVM model showed that, despite the greatest specificity of major outcome in both dataset (training set: 100%; test set: 98.8%) as well as the precision in training set (100%), minor outcomes had the greatest recall (training set: 98.1%; test set: 97.7%) and F-1 score (training set: 92.7%; test set: 92.2%) in both datasets and the precision of 87.2% in test set (Tables 6 and 7). The accuracy of training and test sets were 89.1% and 86.5%, respectively. The confusion matrix showed that the SVM model could successfully identify 1198 cases of minor outcomes, 510 cases of moderate outcomes, and 87 cases of major outcomes in training set and 515 cases of minor outcomes, 206 cases of moderate outcomes, and 27 cases of major outcomes in test datasets (Tables 8 and 9).

Table 6

SVM Characteristics for The Training Set
Labels	Major Effect	Minor Effect	Moderate Effect	Average	Weighted Average
Specificity	1.000000	0.792191	0.959881	0.917358	0.860811
Precision	1.000000	0.878944	0.904255	0.927733	0.894912
Recall	0.690476	0.981967	0.763473	0.811972	0.891261
F1_score	0.816901	0.927604	0.827922	0.857476	0.887616
Accuracy	NaN	NaN	NaN	0.891261	0.891261

Table 7

SVM Characteristics for Test Set
Labels	Major Effect	Minor Effect	Moderate Effect	Average	Weighted Average
Specificity	0.988943	0.777448	0.944541	0.903644	0.846101
Precision	0.750000	0.872881	0.865546	0.829476	0.862761
Recall	0.540000	0.977230	0.717770	0.745000	0.863819
F1_score	0.627907	0.922113	0.784762	0.778261	0.858150
Accuracy	NaN	NaN	NaN	0.865741	0.865741
0.86

Table 8

SVM Confusion Matrix for The Training Set
Prediction	Major Effect	Minor Effect	Moderate Effect
True
Major Effect	87	7	32
Minor Effect	0	1198	22
Moderate Effect	0	158	510
0.89

Table 9

SVM Confusion Matrix for The Training Set
Prediction	Major Effect	Minor Effect	Moderate Effect
True
Major Effect	27	2	21
Minor Effect	1	515	11
Moderate Effect	8	73	206
0.86

This is a retrospective cohort study of NPDS data with the aim of proposing an effective prediction approach to metformin poisoning outcomes. To the best of our knowledge, this is the first study to implement classification approaches in the outcome prediction of metformin poisoning. In the current study, we showed that both DT and SVM algorithms are very effective and powerful tools to predict metformin poisoning outcomes. Moreover, the SVM method predicted more precisely the prognosis of metformin poisoning, compared to the decision tree.

The decision tree and the support vector machine are two of the most well-known machine learning techniques in medicine for uncovering unexplored areas of treatment, prognosis prediction, and diagnosis [15]. One of the noteworthy implications of the decision tree is to establish the intervention for groups of people of varying risk categories because there is a prejudice in terms of selecting high-risk groups to define medical guidelines [16]. Support vector machines, on the other hand, obtain great accuracy when expressing the relationship between multiple elements via a linear feature [17].

In line with other studies, we also showed that acidosis, hypoglycemia, electrolyte abnormality, hypotension, elevated anion gap, elevated creatinine, tachycardia, renal failure, age, and unintentional exposure to metformin are the determinants of metformin poisoning prognosis. As we expected, acidosis and hypoglycemia are the most important factors in determining the outcome. Metformin-associated lactic acidosis (MALA) is a fatal condition that can occur following metformin poisoning and is accompanied by other gastrointestinal symptoms [18]. MALA has been reported to occur at a rate of up to 138 per 100,000 patients per year [19]. It should be noted that early diagnosis of MALA is critical, and it should be suspected in patients with creatinine concentrations of 256 mol/l or higher and lactate concentrations of 8.4 mmol/l or higher [20]. MALA pathogenesis is attributed to a blockage of the complex 1 respiratory chain in mitochondria [21]. Lactic acidosis can cause an increase in anion production, which can lead to an increase in anion gap [22]. Moreover, lactic acidosis is related with gastrointestinal loss, which leads to hypovolemia, renal failure, and increased creatinine secondary to lactic acidosis, which can be recognized following metformin overdose [23]. We believe that this hypovolemia might be the cause of hypotension and compensatory tachycardia amongst these patients. Metformin exposure can also reduce glucose absorption, lower hepatic glucose synthesis, and induce hypoglycemia [24].

In our study, we found that the accuracy of the SVM model to predict the prognosis of metformin poisoning was higher than the DT model. The rationale behind SVM classification is to increase the dimensionality so that separation boundaries may be established more readily, allowing it to surmount the discriminative between variables and observations [25]. As a result, this methodology performs well when the sample size is less than the dimensions. However, if additional dimensions are required, it is critical to include other factors such as the kernel and C function in order to develop a suitable model [26]. Even if the data preparation and interpretation are comprehensible in the decision tree model, it cannot deal with the missing non-leaf node [27].

The strength of our study is that we used broad scale data as well as introduced a classification approach with high accuracy that can be used in clinical practice. However, there are some limitations that should be mentioned. Data collection of NPDS is based on self-reported cases, meaning that every exposure reported to the NPDS might not be a case of poisoning or overdose. Even though NPDS is the largest database of poisoning in the United States, it does not reflect all of the exposure in the country. While many of the important features in metformin poisoning are known, the most important strength of this study is applying machine learning methods to bring and use these features in clinical practice. This study is a benchmark for other studies in this regard.

It is important to note that outcome prediction plays a crucial role in the management of metformin poisoning. Our ML models to predict metformin poisoning outcome was very accurate, which may encourage clinicians to use these algorithms in clinical practice. In line with other studies, our study also showed that acidosis, hypoglycemia, electrolyte abnormality, hypotension, elevated anion gap, elevated creatinine, tachycardia, and renal failure are the most important features to determine the prognosis of metformin. Besides, we found that the support vector model performs more accurately than the decision tree to predict the prognosis. These ML classification models organize our knowledge about metformin poisoning and illustrate the relationship of the important features with each other that can lead to development of practical algorithms to predict the prognosis of metformin poisoning. Using machine learning algorithms in determining metformin poisoning is recommended.

Disclaimers

NPDS is a database which is maintained by the American Association of Poison Control Centers. This database includes the de-identified cases with substance exposure in the United States. However, NPDS does not claim that it includes all of the substance exposure. Moreover, AAPCC does not necessarily confirm all of the NPDS points of view. Also, all of the cases reported to NPDS are not necessarily overdose or toxicity cases.

Acknowledgment

The authors want to thank NPDS for its assistance in the process of providing the data for this study.

Funding

None

Conflict of Interest

The authors declare no conflicts of interest in this study.

Author contributions

OM conceptualized and designed the study. All of authors contributed in writing. Final version of the manuscript was approved by all of the authors.

Consent for publication

Not applicable

Ethics approval

This article does not meet the federal definition of human subject's research.

Data availability

The datasets analyzed during the current study are not publicly available due patients privacy but are available from the corresponding author on reasonable request.

Harding JL, Pavkov ME, Magliano DJ, Shaw JE, Gregg EW. Global trends in diabetes complications: a review of current evidence. Diabetologia. 2019;62(1):3–16.
Ogurtsova K, da Rocha Fernandes JD, Huang Y, Linnenkamp U, Guariguata L, Cho NH, et al. IDF Diabetes Atlas: Global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes research and clinical practice. 2017;128:40–50.
Flory J, Lipska K. Metformin in 2019. JAMA. 2019;321(19):1926–7.
Leonaviciute D, Madsen B, Schmedes A, Buus NH, Rasmussen BS. Severe Metformin Poisoning Successfully Treated with Simultaneous Venovenous Hemofiltration and Prolonged Intermittent Hemodialysis. Case Rep Crit Care. 2018;2018:3868051-.
Lee S-I, Celik S, Logsdon BA, Lundberg SM, Martins TJ, Oehler VG, et al. A machine learning approach to integrate big data for precision medicine in acute myeloid leukemia. Nature Communications. 2018;9(1):42.
Shameer K, Johnson KW, Glicksberg BS, Dudley JT, Sengupta PP. Machine learning in cardiovascular medicine: are we there yet? Heart. 2018;104(14):1156–64.
Chen JH, Asch SM. Machine Learning and Prediction in Medicine - Beyond the Peak of Inflated Expectations. N Engl J Med. 2017;376(26):2507–9.
Johnson KW, Shameer K, Glicksberg BS, Readhead B, Sengupta PP, Björkegren JLM, et al. Enabling Precision Cardiology Through Multiscale Biology and Systems Medicine. JACC: Basic to Translational Science. 2017;2(3):311 – 27.
Podgorelec V, Kokol P, Stiglic B, Rozman I. Decision trees: an overview and their use in medicine. Journal of medical systems. 2002;26(5):445–63.
Yu W, Liu T, Valdez R, Gwinn M, Khoury MJ. Application of support vector machine modeling for prediction of common diseases: the case of diabetes and pre-diabetes. BMC Medical Informatics and Decision Making. 2010;10(1):16.
Amirabadizadeh A, Nakhaee S, Mehrpour O. Risk assessment of elevated blood lead concentrations in the adult population using a decision tree approach. Drug and Chemical Toxicology. 2020:1–8.
Amirabadizadeh A, Nezami H, Vaughn MG, Nakhaee S, Mehrpour O. Identifying Risk Factors for Drug Use in an Iranian Treatment Sample: A Prediction Approach Using Decision Trees. Substance Use & Misuse. 2018;53(6):1030–40.
Liu D, Yu M, Duncan J, Fondario A, Kharrazi H, Nestadt PS. Discovering the Unclassified Suicide Cases Among Undetermined Drug Overdose Deaths Using Machine Learning Techniques. Suicide and Life-Threatening Behavior. 2020;50(2):333–44.
Gummin DD, Mowry JB, Beuhler MC, Spyker DA, Brooks DE, Dibert KW, et al. 2019 Annual Report of the American Association of Poison Control Centers' National Poison Data System (NPDS): 37th Annual Report. Clinical toxicology (Philadelphia, Pa). 2020;58(12):1360 – 541.
Ramezankhani A, Pournik O, Shahrabi J, Azizi F, Hadaegh F, Khalili D. The Impact of Oversampling with SMOTE on the Performance of 3 Classifiers in Prediction of Type 2 Diabetes. Medical Decision Making. 2016;36(1):137–44.
Ramezankhani A, Hadavandi E, Pournik O, Shahrabi J, Azizi F, Hadaegh F. Decision tree-based modelling for identification of potential interactions between type 2 diabetes risk factors: a decade follow-up in a Middle East prospective cohort study. BMJ Open. 2016;6(12):e013336.
Stafford IS, Kellermann M, Mossotto E, Beattie RM, MacArthur BD, Ennis S. A systematic review of the applications of artificial intelligence and machine learning in autoimmune diseases. npj Digital Medicine. 2020;3(1):30.
Goonoo MS, Morris R, Raithatha A, Creagh F. Republished: Metformin-associated lactic acidosis: reinforcing learning points. Drug and Therapeutics Bulletin. 2021;59(8):124–7.
Lalau JD, Kajbaf F, Protti A, Christensen MM, De Broe ME, Wiernsperger N. Metformin-associated lactic acidosis (MALA): Moving towards a new paradigm. Diabetes, obesity & metabolism. 2017;19(11):1502–12.
van Berlo-van de Laar IRF, Gedik A, van ‘t Riet E, de Meijer A, Taxis K, Jansman FGA. Identifying patients with metformin associated lactic acidosis in the emergency department. International Journal of Clinical Pharmacy. 2020;42(5):1286–92.
Owen MR, Doran E, Halestrap AP. Evidence that metformin exerts its anti-diabetic effects through inhibition of complex 1 of the mitochondrial respiratory chain. Biochem J. 2000;348 Pt 3(Pt 3):607–14.
Blough B, Moreland A, Mora A, Jr. Metformin-induced lactic acidosis with emphasis on the anion gap. Proc (Bayl Univ Med Cent). 2015;28(1):31–3.
Arroyo D, Melero R, Panizo N, Goicoechea M, Rodríguez-Benítez P, Vinuesa SG, et al. Metformin-associated acute kidney injury and lactic acidosis. International journal of nephrology. 2011;2011:749653.
Al-Abri SA, Hayashi S, Thoren KL, Olson KR. Metformin overdose-induced hypoglycemia in the absence of other antidiabetic drugs. Clinical toxicology (Philadelphia, Pa). 2013;51(5):444–7.
Verplancke T, Van Looy S, Benoit D, Vansteelandt S, Depuydt P, De Turck F, et al. Support vector machine versus logistic regression modeling for prediction of hospital mortality in critically ill patients with haematological malignancies. BMC medical informatics and decision making. 2008;8:56-.
Yu W, Liu T, Valdez R, Gwinn M, Khoury MJ. Application of support vector machine modeling for prediction of common diseases: the case of diabetes and pre-diabetes. BMC medical informatics and decision making. 2010;10:16-.
Uddin S, Khan A, Hossain ME, Moni MA. Comparing different supervised machine learning algorithms for disease prediction. BMC Medical Informatics and Decision Making. 2019;19(1):281.

No competing interests reported.

Download PDF

Editorial decision: Major revision
21 Apr, 2022
Reviews received at journal
15 Apr, 2022
Reviewers agreed at journal
15 Apr, 2022
Reviewers invited by journal
21 Feb, 2022
Editor assigned by journal
21 Feb, 2022
Editor invited by journal
19 Feb, 2022
Submission checks completed at journal
19 Feb, 2022
First submitted to journal
05 Feb, 2022

You are reading this latest preprint version

Utility Of Support Vector Machine And Decision Tree To Identify The Prognosis Of Metformin Poisoning In The United States: Analysis Of National Poisoning Data System

Status:

Version 1

Abstract

Figures

Background

Methods

Study Population and Eligibility Criteria

Definition Of Terms

Development Of Classification Model

Results

Prognosis Prediction Based on Decision Tree

Prognosis Prediction Based On Support Vector Machine

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1