EarlyDetectionClassifierQcChecker

Early Detection Classifier QC

Liquid Biopsy жидкостная биопсия cfDNA ctDNA CTC exosomes NGS qPCR

Utility description: Early Detection Classifier QC

Early Detection Classifier QC Checker — MCED Classifier Quality Control

ℹ️ Utility performs reliability verification of Multi-Cancer Early Detection classifier predictions according to FDA SaMD guidance and EU IVDR:
• Tissue Confidence: Classifier confidence in predicted tissue of origin.
• CpG Coverage: Depth of target methylation site coverage — basis of epigenomic signal.
• Bisulfite Conversion Rate: Chemical conversion efficiency — incomplete conversion creates systematic artifacts.
• Informative Fragments: Number of methylation-informative fragments — statistical power.
• Background Noise & Feature Completeness: Noise level and feature extraction completeness for model.

⚠️ IMPORTANT:
• MCED tests apply to asymptomatic population — cost of error is extremely high.
• Unreliable tissue prediction may direct patient to wrong diagnostic workup.
• Low bisulfite conversion mimics hypermethylation, creating false-positive signals.

Usage:
EarlyDetectionClassifierQcChecker.exe → demo mode (console output)
EarlyDetectionClassifierQcChecker.exe input.csv output.json → evaluate your data

Input format:
SampleID,ClassifierName,ClassifierVersion,PredictionScore,PredictedTissue,TissueConfidence,MinTissueConfidence,CpG_Coverage_Mean_X,MinCpG_Coverage_X,BisulfiteConversion_Rate,MinBisulfiteConversion_Rate,BackgroundNoise_Score,MaxBackgroundNoise,InformativeFragments_Count,MinInformativeFragments,ArtifactDetected,FeatureCompleteness_Percent,MinFeatureCompleteness

Example:
MCED-001,OncoDetect,v3.2,0.88,Lung,0.92,0.70,45,30,0.995,0.99,0.05,0.15,18500,5000,false,97.5,90

📍 Scope of Application (Usage Where):
• Screening Programs: QC of each MCED test before result delivery.
• Clinical Trials: Prediction validation in multi-center trials.
• MCED Developers: Production classifier performance monitoring.
• Regulatory Inspections: Documentation of SaMD QC system.

— WHY IS THIS NEEDED?
MCED ML classifiers operate at sensitivity boundary where input data quality determines output reliability.
Model may produce confident prediction even on degraded sample — without QC this goes undetected.
Automated verification separates biological signal from technical noise.
This is critically important for patient safety in population screening.

⚠️ CRITICAL:
• Bisulfite Conversion ≥99%: Below this threshold, data is systematically distorted.
• Tissue Confidence ≥0.70: Low confidence = unreliable localization.
• Informative Fragments ≥5000: Lower count reduces statistical power.
• Feature Completeness ≥90%: Missing features may bias prediction.
• Artifacts: Any artifacts require retesting.

Key features:
• Seven-parameter MCED prediction reliability assessment
• Adaptive logic for positive and negative results
• Integration of epigenomic and ML-specific metrics
• Three-tier classification (Reliable / Low Confidence / Technical Fail)
• Compliance with FDA SaMD and EU IVDR requirements

Critical parameters:
• Tissue Confidence: ≥ 0.70 (for positive results)
• CpG Coverage: ≥ 30×
• Bisulfite Conversion: ≥ 99%
• Informative Fragments: ≥ 5,000
• Background Noise: ≤ 0.15
• Feature Completeness: ≥ 90%
• Artifacts: None detected

💡 Usage tips:
1. Confidence Thresholds: Set MinTissueConfidence based on validation cohort.
2. Conversion Control: Include control samples with known methylation status in each run.
3. Drift Monitoring: Track prediction score distribution over time for model drift detection.
4. Retesting: Upon TECHNICAL_FAIL, automatically initiate repeat analysis.
5. Documentation: Save all QC metrics as part of SaMD audit trail.

⚠️ Note: This utility assesses technical reliability of classifier prediction. It does not replace physician clinical interpretation but guarantees that MCED result is based on quality input data and correct model operation.

input.csv

SampleID,ClassifierName,ClassifierVersion,PredictionScore,PredictedTissue,TissueConfidence,MinTissueConfidence,CpG_Coverage_Mean_X,MinCpG_Coverage_X,BisulfiteConversion_Rate,MinBisulfiteConversion_Rate,BackgroundNoise_Score,MaxBackgroundNoise,InformativeFragments_Count,MinInformativeFragments,ArtifactDetected,FeatureCompleteness_Percent,MinFeatureCompleteness
MCED-2026-SCR-001,OncoDetect_MC,v3.2,0.88,Lung,0.92,0.70,45.0,30.0,0.995,0.99,0.05,0.15,18500,5000,false,97.5,90.0
MCED-2026-SCR-002,OncoDetect_MC,v3.2,0.45,Colon,0.52,0.70,22.0,30.0,0.975,0.99,0.22,0.15,3200,5000,true,78.0,90.0
MCED-2026-SCR-003,OncoDetect_MC,v3.2,0.03,None,0.0,0.70,52.0,30.0,0.998,0.99,0.03,0.15,22000,5000,false,98.2,90.0
MCED-2026-SCR-004,OncoDetect_MC,v3.2,0.72,Liver,0.85,0.70,38.0,30.0,0.992,0.99,0.08,0.15,12000,5000,false,94.0,90.0

URS & FS — User Requirements and Functional Specification

This document describes the controlled interface and behaviour of EarlyDetectionClassifierQcChecker for Early Detection Classifier QC Checker.

Domain limits URS input.csv FS IQ/OQ/PQ

Domain limits and critical parameters

Key fragments from the source description are shown below. Before production use, limits must be verified against the approved specification, registration dossier and local SOPs.

This is critically important for patient safety in population screening.
⚠️ CRITICAL:
• Bisulfite Conversion ≥99%: Below this threshold, data is systematically distorted.
• Tissue Confidence ≥0.70: Low confidence = unreliable localization.
• Informative Fragments ≥5000: Lower count reduces statistical power.
• Feature Completeness ≥90%: Missing features may bias prediction.
Critical parameters:
• Tissue Confidence: ≥ 0.70 (for positive results)
• CpG Coverage: ≥ 30×
• Bisulfite Conversion: ≥ 99%
• Informative Fragments: ≥ 5,000
• Background Noise: ≤ 0.15
• Feature Completeness: ≥ 90%

URS — User Requirements Specification

ID	Requirement	Criticality	Acceptance criterion
URS-001	The utility shall accept an input.csv file for Early Detection Classifier QC Checker with headers defined in the data contract.	High	The file is processed without manual header editing.
URS-002	The utility shall perform deterministic QC evaluation without machine learning and without probabilistic conformance decisions.	High	Identical input data, rule version and configuration produce reproducible results.
URS-003	The utility shall validate mandatory fields, data types, ranges, units and domain plausibility.	High	Schema, conversion and range errors are explicitly reported.
URS-004	The utility shall apply domain limits and rules from the description, approved specification, registration dossier and local SOPs.	High	Each check has PASS/WARNING/FAIL and a clear message.
URS-005	The utility shall generate output.json with machine-readable results, source values, warnings, failures and critical findings.	High	JSON is suitable for LIMS/ELN/MES integration and QA/QC review.
URS-006	The utility shall preserve traceability between batch/sample, input file, applied rules and final status.	High	Output contains identifiers, checked parameters and audit metadata.
URS-007	The documentation shall support IQ/OQ/PQ, CSV/CSA and review by internal QA or inspectors.	Medium	URS, FS, input/output contract and test scenarios are supplied with the utility.
URS-008	The utility shall be used as a QC decision-support tool and not as a substitute for approved specifications and QA/QP release decision.	Medium	Documentation states change control and limit-verification expectations.

input.csv contract

#	Field	Type	Sample	Purpose
1	`SampleID`	string / controlled vocabulary	`MCED-2026-SCR-001`	Sample or laboratory specimen identifier.
2	`ClassifierName`	string / controlled vocabulary	`OncoDetect_MC`	Biological/molecular component controlled as a CQA.
3	`ClassifierVersion`	string / controlled vocabulary	`v3.2`	Controlled input parameter for deterministic QC rules.
4	`PredictionScore`	decimal	`0.88`	Controlled input parameter for deterministic QC rules.
5	`PredictedTissue`	string / controlled vocabulary	`Lung`	Controlled input parameter for deterministic QC rules.
6	`TissueConfidence`	string / controlled vocabulary	`0.92`	Controlled input parameter for deterministic QC rules.
7	`MinTissueConfidence`	string / controlled vocabulary	`0.70`	Controlled input parameter for deterministic QC rules.
8	`CpG_Coverage_Mean_X`	decimal	`45.0`	Controlled input parameter for deterministic QC rules.
9	`MinCpG_Coverage_X`	decimal	`30.0`	Controlled input parameter for deterministic QC rules.
10	`BisulfiteConversion_Rate`	decimal	`0.995`	Controlled input parameter for deterministic QC rules.
11	`MinBisulfiteConversion_Rate`	decimal	`0.99`	Controlled input parameter for deterministic QC rules.
12	`BackgroundNoise_Score`	decimal	`0.05`	Controlled input parameter for deterministic QC rules.
13	`MaxBackgroundNoise`	decimal	`0.15`	Controlled input parameter for deterministic QC rules.
14	`InformativeFragments_Count`	integer / decimal	`18500`	Count parameter used for microbiological, particulate or cellular control.
15	`MinInformativeFragments`	decimal	`5000`	Controlled input parameter for deterministic QC rules.
16	`ArtifactDetected`	string / controlled vocabulary	`false`	Controlled input parameter for deterministic QC rules.
17	`FeatureCompleteness_Percent`	decimal	`97.5`	Controlled input parameter for deterministic QC rules.
18	`MinFeatureCompleteness`	decimal	`90.0`	Controlled input parameter for deterministic QC rules.

SampleID,ClassifierName,ClassifierVersion,PredictionScore,PredictedTissue,TissueConfidence,MinTissueConfidence,CpG_Coverage_Mean_X,MinCpG_Coverage_X,BisulfiteConversion_Rate,MinBisulfiteConversion_Rate,BackgroundNoise_Score,MaxBackgroundNoise,InformativeFragments_Count,MinInformativeFragments,ArtifactDetected,FeatureCompleteness_Percent,MinFeatureCompleteness
MCED-2026-SCR-001,OncoDetect_MC,v3.2,0.88,Lung,0.92,0.70,45.0,30.0,0.995,0.99,0.05,0.15,18500,5000,false,97.5,90.0
MCED-2026-SCR-002,OncoDetect_MC,v3.2,0.45,Colon,0.52,0.70,22.0,30.0,0.975,0.99,0.22,0.15,3200,5000,true,78.0,90.0
MCED-2026-SCR-003,OncoDetect_MC,v3.2,0.03,None,0.0,0.70,52.0,30.0,0.998,0.99,0.03,0.15,22000,5000,false,98.2,90.0

Input validation rules

ID	Field	Rule	Criticality
VR-001	`SampleID`	The field shall match an approved dictionary or accepted string representation.	High
VR-002	`ClassifierName`	The field shall match an approved dictionary or accepted string representation.	High
VR-003	`ClassifierVersion`	The field shall match an approved dictionary or accepted string representation.	High
VR-004	`PredictionScore`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-005	`PredictedTissue`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-006	`TissueConfidence`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-007	`MinTissueConfidence`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-008	`CpG_Coverage_Mean_X`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-009	`MinCpG_Coverage_X`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-010	`BisulfiteConversion_Rate`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-011	`MinBisulfiteConversion_Rate`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-012	`BackgroundNoise_Score`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-013	`MaxBackgroundNoise`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-014	`InformativeFragments_Count`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-015	`MinInformativeFragments`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-016	`ArtifactDetected`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-017	`FeatureCompleteness_Percent`	The field shall match an approved dictionary or accepted string representation.	Medium
VR-018	`MinFeatureCompleteness`	The field shall match an approved dictionary or accepted string representation.	Medium

FS — Functional Specification

ID	Function	Implementation
FS-001	CLI execution	Support execution modes: demo mode without arguments and production mode input.csv output.json.
FS-002	CSV import	Read input.csv in UTF-8/CSV-compatible format and validate header and expected columns.
FS-003	Schema validation	Check mandatory fields, column count, unknown key fields and empty mandatory values.
FS-004	Type conversion	Convert numeric, flag and text values; invalid format is recorded as a row-level error.
FS-005	Domain rule engine	Apply rules for Early Detection Classifier QC Checker, including critical limits from the description and approved specification.
FS-006	Status aggregation	Produce final status: FAIL for critical failure, WARNING for non-critical deviation, PASS for conformance.
FS-007	JSON export	Write output.json with detailed checks, source values, warnings, failures and critical findings.
FS-008	Audit support	Keep result structure suitable for review, deviation investigation and calculation reproduction.
FS-009	Integration contract	Support the scenario LIMS/ELN/MES → input.csv → utility → output.json → portal/admin review.
FS-010	Error handling	Return explicit messages for missing file, empty CSV, invalid schema, output write failure and invalid format.

Example output.json

{
  "utilityId": "earlydetectionclassifierqcchecker",
  "utilityFolder": "EarlyDetectionClassifierQcChecker",
  "package": "LiquidBiopsy",
  "overallStatus": "PASS|WARNING|FAIL",
  "sourceFile": "input.csv",
  "processedAtUtc": "2026-06-10T00:00:00Z",
  "checks": [
    {
      "parameter": "SampleID",
      "value": "MCED-2026-SCR-001",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-001"
    },
    {
      "parameter": "ClassifierName",
      "value": "OncoDetect_MC",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-002"
    },
    {
      "parameter": "ClassifierVersion",
      "value": "v3.2",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-003"
    },
    {
      "parameter": "PredictionScore",
      "value": "0.88",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-004"
    },
    {
      "parameter": "PredictedTissue",
      "value": "Lung",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-005"
    },
    {
      "parameter": "TissueConfidence",
      "value": "0.92",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-006"
    },
    {
      "parameter": "MinTissueConfidence",
      "value": "0.70",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-007"
    },
    {
      "parameter": "CpG_Coverage_Mean_X",
      "value": "45.0",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-008"
    },
    {
      "parameter": "MinCpG_Coverage_X",
      "value": "30.0",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-009"
    },
    {
      "parameter": "BisulfiteConversion_Rate",
      "value": "0.995",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-010"
    },
    {
      "parameter": "MinBisulfiteConversion_Rate",
      "value": "0.99",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-011"
    },
    {
      "parameter": "BackgroundNoise_Score",
      "value": "0.05",
      "status": "PASS|WARNING|FAIL",
      "message": "Deterministic rule-based check result",
      "ruleReference": "FS-RULE-012"
    }
  ],
  "criticalFindings": [],
  "warnings": [],
  "audit": {
    "inputHash": "sha256:<calculated at runtime>",
    "rulesVersion": "<utility executable version>",
    "documentation": "EarlyDetectionClassifierQcChecker.documentation.html"
  }
}

Traceability matrix

URS	FS	Test	Evidence
URS-001	FS-001, FS-002	OQ-001	Verify execution and import of valid input.csv.
URS-002	FS-005, FS-006	OQ-004	Repeat the same dataset and compare output.json.
URS-003	FS-003, FS-004, FS-010	OQ-002, OQ-003	Verify missing columns and invalid types.
URS-004	FS-005, FS-006	OQ-004, PQ-001	Verify critical deviations on real/boundary data.
URS-005	FS-007, FS-009	OQ-005	Verify JSON schema and downstream-system suitability.
URS-006	FS-008	OQ-006	Verify identifiers and audit metadata.
URS-007	FS-008, FS-010	IQ-001, OQ-007	Verify documentation completeness and control evidence.
URS-008	FS-005, FS-008	PQ-002	Verify review workflow and no replacement of QA decision.

IQ/OQ/PQ test scenarios

ID	Scenario	Expected result
IQ-001	Verify executable, input.csv, documentation and checksum availability.	Delivery set is complete; version is recorded.
OQ-001	Valid sample row from input.csv.	PASS or acceptable WARNING according to rules.
OQ-002	Remove a mandatory CSV column.	Schema error or FAIL with missing-column reference.
OQ-003	Place a non-numeric value into a numeric field.	Type-conversion error with row/field reference.
OQ-004	Set a critical parameter outside the limit.	FAIL and critical finding.
OQ-005	Verify output.json structure.	All mandatory sections are present and JSON is valid.
OQ-006	Verify batch/sample traceability.	Input and result identifiers match.
PQ-001	Verify 3–5 real user batches/samples.	Result is confirmed by QC/QA review.
PQ-002	Verify deviation workflow and manual QA decision.	Utility supports review but does not replace approved decision.

QA/QC and change control

Do not rename columns without updating validator, documentation and test set.
Retain input.csv, output.json, executable version and checksum.
Before production use, perform IQ/OQ/PQ or equivalent CSV/CSA verification.
Critical limits shall be verified against the approved specification, registration dossier and local SOPs.
The utility provides structured QC decision support; final release decision remains with QA/QP and approved procedures.

Included in packages

Liquid Biopsy QC Suite

QC and pre-analytical control package for liquid biopsy workflows: cfDNA/ctDNA, CTC, EV/exosomes, methylation, NGS/qPCR/ddPCR, sample quality, contamination, sensitivity and reporting checks.

Open