Pith. sign in

REVIEW 3 major objections 5 minor 13 references

StackLiverNet: A Novel Stacked Ensemble Model for Accurate and Interpretable Liver Disease Detection

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that a five-biomarker stacked ensemble of XGBoost and k-nearest neighbors with a LightGBM meta-model detects liver disease with 99.89% held-out accuracy and per-patient explanations.

desk verdict A textbook data-leakage error – feature selection and scaling before the split – invalidates the headline 99.89% test accuracy, so this stacking ensemble is not what it claims. read the letter →

arxiv 2508.00117 v2 pith:MGZLOQDR submitted 2025-07-31 cs.LG cs.AI

Md. Ehsanul Haque , S. M. Jahidul Islam , Shakil Mia , Rumana Sharmin , Ashikuzzaman , Md Samir Morshed , Md. Tahmidul Huque This is my paper · ORCID
classification cs.LGcs.AI
keywords liverdiseaseclassificationStackNetstackedensemblelearningexplainableAILIMESHAPfeatureselectionclassimbalance
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper aims to establish that liver disease can be detected almost perfectly from routine blood-test values by a stacked ensemble it calls StackLiverNet, and that the model is fast and transparent enough for clinical screening. The authors train XGBoost and k-nearest neighbors as base classifiers, feed their predictions to a LightGBM meta-model, and report 99.89% accuracy, 0.9974 Cohen's kappa, and 0.9993 AUC on a held-out test set, with only five misclassified samples. They also report that the entire pipeline trains in about 4.3 seconds and classifies a batch in about 0.11 seconds. The practical stake is that, if the estimate is unbiased, a cheap five-feature blood panel could flag likely liver disease with near-perfect accuracy while explaining each decision in terms clinicians can review.

What carries the argument

The carrying object is StackLiverNet itself, a two-level stacking ensemble: the predictions of XGBoost and KNN on each training sample are concatenated into a new two-column feature matrix, and a LightGBM classifier is trained on that matrix to produce the final label. Around this core sits a preprocessing chain that does much of the work: mode imputation and Winsorization tame missing values and outliers, one-way ANOVA drops the two non-significant features, recursive feature elimination with cross-validation keeps five lab biomarkers, and random undersampling equalizes the training classes. The ensemble mechanism is what lets the model combine complementary decision patterns, while the preprocessing chain is what keeps the learning signal clean.

What would settle it

Run the exact StackLiverNet pipeline again, but refit the ANOVA threshold, the RFE-CV selector, and the standard scaler inside each cross-validation fold on the training fold only, then apply the fitted transformations to the held-out fold. If the resulting test accuracy falls well below 99.89%, the reported figure was inflated by data leakage; if it stays near 99.9%, the claim survives.

Watch

Extended reading notes

Core claim

In the paper's own terms, the central discovery is that a stacking ensemble built from just two diverse base learners—XGBoost and k-nearest neighbors—combined by a LightGBM meta-classifier outperforms individual gradient-boosting and neural-network baselines on the liver patient dataset. After one-way ANOVA removes Gender and Age as non-significant, recursive feature elimination with cross-validation retains five features—Total Bilirubin, Alkaline Phosphatase, Aspartate Aminotransferase, Alanine Aminotransferase, and Albumin—and random undersampling balances the training classes. The resulting StackLiverNet reaches a mean 5-fold cross-validation accuracy of 99.95% and a test accuracy of 99.89% with AUC 0.9993, with LIME, SHAP, and Morris sensitivity analysis all pointing to alkaline phosphatase, bilirubin, and the transaminases as the decision-driving biomarkers. The authors interpret these results as evidence that the model is accurate, stable, and explainable enough to support early liver-disease screening.

Load-bearing premise

The headline accuracy depends on the assumption that every preprocessing step that touches the target variable—especially the ANOVA filter and RFE-CV feature selection—was fitted only on the training part and never saw the test split before the final evaluation.

Editorial extensions

If this is right

  • If StackLiverNet's estimate is unbiased, a five-biomarker blood panel—total bilirubin, alkaline phosphatase, AST, ALT, and albumin—is enough to separate liver disease from non-disease in this patient population with near-perfect accuracy.
  • The reported training time of 4.28 seconds and inference time of 0.11 seconds imply the full pipeline can run on modest hardware without delaying a clinical screening workflow.
  • The LIME and SHAP findings tie the model's decisions to known liver biomarkers, which means a clinician can audit individual predictions rather than treating the model as a black box.
  • Because the stacking design beats both XGBoost and KNN individually, the result suggests that complementary base learners plus a small meta-model can outperform either family alone on this task.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An inference the paper leaves implicit: the near-perfect separation on five routine lab values suggests the dataset's classes are strongly separable, so the accuracy ceiling is not evidence about harder real-world populations with similar lab panels; an independent hospital cohort is needed.
  • A caution implied by the reported pipeline: because feature selection and scaling appear to be applied before the single 80/20 split, the headline accuracy is best treated as an estimate that should be rechecked with selection nested inside cross-validation.
  • A testable extension: comparing StackLiverNet against the same five features with no feature selection at all would show how much of the gain comes from the ensemble versus the preprocessing.
  • A neighboring problem to watch: the same stacking recipe—two diverse base learners plus a fast gradient-boosting meta-model—transfers naturally to other tabular medical screening tasks where interpretability matters.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes StackLiverNet, a stacked ensemble (XGBoost and KNN base classifiers with a LightGBM meta-model) for binary liver disease detection on a Kaggle clinical dataset. The pipeline includes missing-value handling, Winsorization, standardization, one-way ANOVA feature filtering, RFE-CV feature selection, random undersampling, hyperparameter-tuned base models, and post-hoc interpretability with LIME, SHAP, and the Morris method. The central claim is near-perfect held-out performance: 99.89% test accuracy, Cohen's Kappa 0.9974, AUC 0.9993, and only 5 misclassifications on a test set of 4,674 samples.

Significance. If the reported results were valid, the contribution would be practically attractive: a fast, interpretable model that reaches essentially perfect discrimination from five clinical laboratory features, with training time around 4 seconds and inference around 0.1 seconds. The paper also integrates three XAI techniques, which is a strength for clinical acceptability. However, the central numerical claim is undermined by the described evaluation protocol: feature selection and standardization are performed on the entire dataset before the train/test split, so the reported test accuracy is not an unbiased generalization estimate. The paper provides no code or data artifacts to allow independent verification. Because the headline result is the main contribution, the validity concern is load-bearing.

major comments (3)
  1. [III-B, III-C, III-D, III-E] The experimental pipeline leaks test information into model construction. Section III-B standardizes all numerical features, Section III-C performs one-way ANOVA filtering, and Section III-D runs RFE-CV feature selection on the full dataset, all before the 80/20 split described in Section III-E. The test split therefore contributes to deciding which five features are retained and to the values of the scaling parameters. The reported 99.89% test accuracy and the 'only 5 misclassifications' claim are consequently not unbiased estimates of generalization. The paper does not state that feature selection or standardization was refit inside the training folds, and Algorithm 1, which operates on the full X and y, makes no provision for a split. This leakage alone invalidates the paper's central empirical claim.
  2. [III-A and Table II] The data-set description and the class-frequency tables do not reconcile. Section III-A states that the dataset contains 26,303 samples (16,749 liver patients). Table II implies 13,414 + 5,282 = 18,696 samples in the original training set and 3,335 + 1,339 = 4,674 samples in the test set, for a total of 23,370 samples. This leaves 2,933 samples unaccounted for, and no post-cleaning sample count or deduplication step is reported. The cited Kaggle dataset is known to contain duplicate rows; if duplicates are present and not removed, identical or near-identical records may appear in both training and test partitions, further inflating the held-out accuracy. The manuscript must report the exact sample size after each preprocessing stage and explain the discrepancy.
  3. [IV, Table IV] The claim that StackLiverNet is 'slightly but significantly better' than the baseline models is not supported by the reported numbers or by any significance test. Table IV shows StackLiverNet test AUC 0.9993, which is lower than XGBoost's 0.9995 and essentially tied with KNN's 0.9992; the test accuracy margin over KNN is 0.0008. No confidence intervals, statistical tests, or repeated-split analyses are provided. Given the feature-selection leakage described above, the comparative superiority claim is also not valid. The authors should either provide a proper statistical comparison under a leak-free protocol or temper the claim.
minor comments (5)
  1. [III-G, Algorithm 2] Algorithm 2 is incomplete: the pseudocode ends abruptly after the meta-training-set construction and does not show the actual training of the meta-model or the prediction step on the test set, and the line numbering with '0:' prefixes is nonstandard.
  2. [Table VII] The header for the last column appears as a stray 'µ*' and is not defined; the table also reports a confidence interval column that is never explained in the text. The values for SGOT and ALB (µ* = 0.0008 and 0.0000) sit oddly with the LIME discussion in Section IV-B, which describes SGOT as an important feature; the authors should reconcile the sensitivity results with the local explanations.
  3. [Figure 3] The caption contains typos: 'eft' should be 'left' and one panel label is truncated as 'Curve (Right)'. Please also number the subfigures consistently.
  4. [References] Reference formatting is inconsistent (e.g., [4] is a self-citation to a different conference paper, [3] is a preprint without a DOI, and several entries mix capitalization styles). The authors should verify that all references are relevant and complete.
  5. [Throughout] There are numerous typographical and grammatical issues (e.g., 'Cohen s Kappa', 'ModelAnagnostic', 'spliting') and inconsistent use of variable names (e.g., 'Alkphos' vs. 'ALP' vs. 'Alkaline Phosphatase'). A careful proofreading pass is needed.

Circularity Check

1 steps flagged · score 4.0 of 10

Pre-split ANOVA filtering and RFE-CV feature selection leak test-set information into the model, so the abstract's 99.89% test accuracy is partly a fitted quantity rather than an unbiased held-out prediction.

  1. fitted input called prediction [Sections III-B (standardization), III-C (ANOVA), III-D (RFE-CV), III-E (80/20 split); abstract's 'testing accuracy of 99.89%']
    "A one-way ANOVA was performed to determine the significance of each feature against the target variable where a significance level of 0.05 was used (95% confidence) [12]. ... these two characteristics were eliminated as dataset to be further analyzed. ... The Recursive Feature Elimination with Cross-Validation (RFE-CV) was used to determine the most pertinent features to be used by the model [13]. ... The dataset was split into training and testing sets using an 80-20 ratio."

    By the paper's own ordering, standardization (III-B), the ANOVA filter (III-C), and RFE-CV selection (III-D) all run on the full dataset before 'The dataset was split into training and testing sets using an 80-20 ratio' (III-E). Test rows therefore help decide which 5 features are retained and set the scaling parameters, so the abstract's 'testing accuracy of 99.89%' is not an information-free held-out estimate; the feature set and scaler are partly fitted on the test data. The paper never states that these steps were refit inside training folds only, and no code or data artifact is provided to rule out the leakage.

full rationale

I walked the paper's derivation chain from data collection (III-A) through preprocessing (III-B), ANOVA (III-C), RFE-CV (III-D), the 80/20 split (III-E), baseline training (III-F), stacking (III-G), and interpretation (III-H, IV-B). The only load-bearing circularity is the pre-split feature-selection and scaling step: the test split is created after standardization, ANOVA filtering, and RFE-CV have already used the entire dataset, so test information influences the feature set and scaling, making the reported held-out accuracy a partially fitted quantity (pattern: fitted input called prediction). This is a moderate, not full, circularity since the base and meta-model parameters are fit only to training labels and the reported metrics are genuine computations on the retained features. No self-definitional equivalence, no load-bearing self-citation (ref [4], the authors' prior CKD paper, is cited only as background for class-imbalance limitations), no uniqueness theorem, no ansatz smuggled via citation, and no renaming of a known result were found. A separate reporting gap, not a circularity: Section III-A states 26,303 samples with 16,749 liver patients, while Table II implies 23,370 rows remain after cleaning, and the paper never reports the post-cleaning sample count or any deduplication, so duplicate-row leakage cannot be excluded. Because the central classification claim retains independent content (near-perfect separation on the chosen 5 clinical features, with plausible timings), the score is 4 rather than 6+.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The model has no invented physical entities. The free parameters are standard ML hyperparameters and data-processing choices, several of which were tuned on the same data that later produced the reported test metrics. The key assumptions are dataset representativeness, missing-data ignorability, and the validity of the preprocessing sequence.

free parameters (7)
  • Number of features selected by RFE-CV = 5
    Chosen because CV scores were similar for 5 and 7 features; selection used the full dataset before splitting.
  • Random undersampling ratio = 1:1 (5,282 disease, 5,282 no-disease in training)
    Majority class reduced to match the minority class; this changes the class prior and affects reported accuracy.
  • CatBoost hyperparameters = iterations=150, depth=6, learning rate=0.1, l2_leaf_reg=3
    Selected by initial experiments and cross-validation on the training data (Table III).
  • MLP hyperparameters = hidden layer sizes=(100,50), relu, adam, alpha=0.0001, max_iter=300
    Hand-tuned, reported in Table III.
  • KNN hyperparameters = n_neighbors=7, weights=distance, p=2
    Hand-tuned, reported in Table III.
  • XGBoost hyperparameters = n_estimators=150, max_depth=5, learning rate=0.1, subsample=0.8, colsample_bytree=0.8
    Hand-tuned, reported in Table III.
  • Standardization mean and standard deviation = not specified
    Applied to all numerical features in preprocessing (Section III-B); whether statistics were computed on the training set only is not stated.
assumptions (5)
  • domain assumption The Kaggle Liver Disease Patient Dataset labels are clinically accurate and representative of the target population.
    All predictive conclusions depend on dataset quality; the paper provides no external clinical validation.
  • domain assumption Deleting rows with missing values does not introduce selection bias.
    Section III-B removes missing rows silently; class counts before and after suggest about 2,933 no-disease rows were removed, which alters the class prior.
  • ad hoc to paper One-way ANOVA p-value below 0.05 identifies features useful for classification.
    Section III-C uses ANOVA to drop Age and Gender; statistical association does not guarantee predictive relevance.
  • domain assumption RFE-CV with Random Forest feature importances finds a near-optimal feature subset.
    Section III-D uses random forest importance to rank features; this is a standard heuristic, not an optimality guarantee.
  • domain assumption Undersampling the majority class during training does not distort the decision boundary relative to the original population.
    Section III-E rebalances training to 5,282/5,282, while the test set remains imbalanced; the effect on calibration and generalization is not analyzed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of StackLiverNet: A Novel Stacked Ensemble Model for Accurate and Interpretable Liver Disease Detection." pith.science (2026). https://pith.science/paper/MGZLOQDR

@misc{pith2026250800117,
  author       = {Pith},
  title        = {Pith review of: StackLiverNet: A Novel Stacked Ensemble Model for Accurate and Interpretable Liver Disease Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MGZLOQDR}},
  note         = {Machine review of arXiv:2508.00117}
}
read the original abstract

Liver diseases are a serious health concern in the world, which requires precise and timely diagnosis to enhance the survival chances of patients. The current literature implemented numerous machine learning and deep learning models to classify liver diseases, but most of them had some issues like high misclassification error, poor interpretability, prohibitive computational expense, and lack of good preprocessing strategies. In order to address these drawbacks, we introduced StackLiverNet in this study; an interpretable stacked ensemble model tailored to the liver disease detection task. The framework uses advanced data preprocessing and feature selection technique to increase model robustness and predictive ability. Random undersampling is performed to deal with class imbalance and make the training balanced. StackLiverNet is an ensemble of several hyperparameter-optimized base classifiers, whose complementary advantages are used through a LightGBM meta-model. The provided model demonstrates excellent performance, with the testing accuracy of 99.89%, Cohen Kappa of 0.9974, and AUC of 0.9993, having only 5 misclassifications, and efficient training and inference speeds that are amenable to clinical practice (training time 4.2783 seconds, inference time 0.1106 seconds). Besides, Local Interpretable Model-Agnostic Explanations (LIME) are applied to generate transparent explanations of individual predictions, revealing high concentrations of Alkaline Phosphatase and moderate SGOT as important observations of liver disease. Also, SHAP was used to rank features by their global contribution to predictions, while the Morris method confirmed the most influential features through sensitivity analysis.

Figures

Figures reproduced from arXiv: 2508.00117 by the authors.

Figure 2
Figure 2. Cross-validation scores for different numbers of features during Recursive Feature Elimination with Cross￾Validation (RFE-CV). Algorithm 1 RFECV with Random Forest Require: Features X, labels y, folds k = 5 Ensure: Optimal features F∗, accuracy list A, accuracies at 3, 5, 7 features 0: F ← X; A ← []; S ← {} 0: while |F| > 1 do 0: Train RF with k-fold CV on F, get mean accuracy a 0: Append a to A; if |F| ∈ {3, 5, 7},… view at source ↗
Figure 3
Figure 3. Comprehensive evaluation results of the StackLiverNet model: confusion matrix (eft), Validation Curve (Middle), ROC [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. LIME explanations showing the most influential fea [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

13 extracted references · 12 canonical work pages

  1. [1]

    Liver disease in numbers – key facts and statistics,

    British Liver Trust, “Liver disease in numbers – key facts and statistics,” 2024, accessed: 2025-06-14. [Online]. Available: https://britishlivertrust.org.uk/information-and-support/statistics/#:∼: text=Together%20alcohol%2C%20viral%20hepatitis%2C%20and,Pre% 2Dpandemic%20estimates%20suggest

    work page 2024

  2. [2]

    Diagnosis and treatment of liver disease: Current trends and future directions,

    H. Wazir, M. Abid, B. Essani, H. Saeed, M. Ahmad Khan, F. Nasrullah, U. Qadeer, A. Khalid, G. Varrassi, M. A. Muzammil, A. Maryam, A. R. S. Syed, A. A. Shah, S. Kinger, and F. Ullah, “Diagnosis and treatment of liver disease: Current trends and future directions,” Cureus, vol. 15, no. 12, p. e49920, 2023. [Online]. Available: https://doi.org/10.7759/cureus.49920

    work page doi:10.7759/cureus.49920 2023

  3. [3]

    Unveiling explainable ai in healthcare: Current trends, challenges, and future directions,

    N. A. Aziz, A. Manzoor, M. D. M. Qureshi, M. A. Qureshi, and W. Rashwan, “Unveiling explainable ai in healthcare: Current trends, challenges, and future directions,” medRxiv, 2024, preprint. [Online]. Available: https://www.medrxiv.org/content/10.1101/2024.08. 10.24311735v2.full

    doi:10.1101/2024.08 2024

  4. [4]

    Improving chronic kidney disease detection efficiency: Fine tuned catboost and nature-inspired algorithms with explainable ai,

    M. E. Haque, S. M. J. Islam, J. Maliha, M. S. H. Sumon, R. Sharmin, and S. Rokoni, “Improving chronic kidney disease detection efficiency: Fine tuned catboost and nature-inspired algorithms with explainable ai,” in 2025 IEEE 14th International Conference on Communication Systems and Network Technologies (CSNT), 2025, pp. 811–818

    work page 2025

  5. [5]

    Improved liver disease prediction from clinical data through an evaluation of ensemble learning approaches,

    S. M. Ganie, P. K. Dutta Pramanik, and Z. Zhao, “Improved liver disease prediction from clinical data through an evaluation of ensemble learning approaches,” BMC Medical Informatics and Decision Making , vol. 24, no. 1, p. 160, 2024

    work page 2024

  6. [6]

    Adaptive method for exploring deep learning tech - niques for subtyping and prediction of liver disease,

    A. M. Hendi, M. A. Hossain, N. A. Majrashi, S. Limkar, B. M. Elamin, and M. Rahman, “Adaptive method for exploring deep learning tech - niques for subtyping and prediction of liver disease,” Applied Sciences, vol. 14, no. 4, p. 1488, 2024

    work page 2024

  7. [7]

    Explainability enhanced liver disease diagnosis technique using tree selection and stacking ensemble -based random forest model,

    M. R. Khatun, M. Saifuddin, M. R. Jani, M. R. Islam, M. R. Chowdhury, and M. A. Rahman, “Explainability enhanced liver disease diagnosis technique using tree selection and stacking ensemble -based random forest model,” Informatics and Health, vol. 2, pp. 17–40, 2025

    work page 2025

  8. [8]

    Performance analysis of machine learning models for liver disease patient classification,

    A. K. Jena, A. Patro, I. Behera, and S. Rath, “Performance analysis of machine learning models for liver disease patient classification,” in Proceedings of the 2024 International Conference on Intelligent Computing and Sustainable Innovations in Technology (IC-SIT), 2024

    work page 2024

Show all 13 references
  1. [9]

    A comparative study of machine learning algorithms using explainable artificial intelligence system for predicting liver disease,

    A. Nilofer and S. Sasikala, “A comparative study of machine learning algorithms using explainable artificial intelligence system for predicting liver disease,” Computing Open, 2024

    2024

  2. [10]

    Prediction of chronic liver disease patients using integrated projection based statistical feature extraction with machine learning algorithms,

    R. Amin, R. Yasmin, S. Ruhi, M. H. Rahman, and M. S. Reza, “Prediction of chronic liver disease patients using integrated projection based statistical feature extraction with machine learning algorithms,” Informatics in Medicine Unlocked, vol. 36, p. 101155, 2023. [Online]. Av...

    2023

  3. [11]

    Liver disease patient dataset,

    A. Shrivastava, “Liver disease patient dataset,” https://www.kaggle.com/ datasets/abhi8923shriv/liver-disease-patient-dataset, 2020, accessed: 2025-05-29

    2020

  4. [12]

    Analysis of variance (anova),

    L. Sta˚hle and S. Wold, “Analysis of variance (anova),” Chemometrics and Intelligent Laboratory Systems, vol. 6, no. 4, pp. 259–272, 1989

    1989

  5. [13]

    Recursive feature elimination with cross - validation with decision tree: Feature selection method for machine learning-based intrusion detection systems,

    M. Awad and S. Fraihat, “Recursive feature elimination with cross - validation with decision tree: Feature selection method for machine learning-based intrusion detection systems,” Journal of Sensor and Actuator Networks, vol. 12, p. 67, 09 2023

    2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.