The Central Sterile Supply Department (CSSD) is a critical hospital unit responsible for the reprocessing of reusable surgical instruments. Although surgical instruments undergo automated washing and disinfection cycles, they may appear clean to the naked eye while still harbouring microscopic contaminants that pose a risk to patient safety.

At present, reliance on manual inspection processes – particularly during post-disinfection inspection and final packing stations – introduces significant operational and clinical risks. Relying on human sight to verify hundreds of complex, often similar instruments can result in the following challenges:

a. Patient Safety Risks: Undetected instrument damage, contamination, or shortages may compromise patient safety and clinical outcomes.

b. Operational Inefficiency: Labour-intensive manual inspection processes prolongs processing times and may contribute to staff fatigue and burnout.

c. Workflow Disruptions: Inconsistent instrument readiness can delay surgeries procedures and forces schedule reshuffles.

d. Limited Data Visibility: Poor data capture hinders the ability to perform preventative maintenance, hiking inventory and reprocessing costs.

e. Compliance and Traceability Gaps: Error-prone manual tracking hinders regulatory auditability, weakens instrument lifecycle traceability and increases compliance risks.

To address these challenges, Woodlands Hospital (WH) is seeking innovative AI-enabled solutions that can enhance quality assurance, improve instrument traceability, and increase operational efficiency within the CSSD workflow.

Outlined below is the strict CSSD workflow for the handling of contaminated surgical instruments upon receipt. The two highlighted process steps represent the key focus areas of this Call for Innovation.

The proposed solution(s) should deliver three core capabilities that collectively enhance accuracy, efficiency, and quality assurance within CSSD operations.

1. Instrument Identification & Counting

The solution should accurately identify and count surgical instruments across different shapes, sizes, and manufacturers, with a target accuracy of at least 95%.

i) Scaling Differences: The solution must intelligently account for variations in dimensions and design by focusing on proportional characteristics rather than fixed measurements. Examples:

(1) A "medium" haemostat from different vendors may range from 125mm - 135mm in length.

(2) Surgical scissors labelled as "5-inch" may actually measure between 4.8 - 5.2 inches.

(3) Forceps designated as "standard" may have jaw widths varying by 2 - 3mm.

ii) Multi-Vendor Compatibility: The solution should support multi-vendor compatibility through adaptive learning, enabling seamless onboarding of new instrument types. Examples:

(1) Clamps: Ratchet mechanisms may have different tooth patterns, jaw shapes, or locking mechanisms while maintaining the same clinical purpose.

(2) Forceps: Vendor A may produce forceps with straight tips, while Vendor B's equivalent has slightly curved tips, yet both serve the same surgical function.

iii) Discrepancy & Lifecycle Alerts: The solution should support real-time discrepancy detection and lifecycle tracking should also be incorporated to flag incomplete sets, detect abnormalities, and support predictive maintenance insights. Examples:

(1) Maintenance History: Some instruments may have been refurbished, re-sharpened, or had components replaced.

(2) Usage-Related Changes: Frequently used instruments may have worn markings, smoothed textures, or slightly altered profiles.

2. Post-Disinfection Quality Check

(i) Advanced Contamination Detection: The solution should enable advanced contamination detection using multi-spectral imaging to identify microscopic residues (e.g. blood, protein, residual organic matter) that are not visible to the human eye. This includes detecting contamination in complex instrument structures such as joints and lumens. Examples:

(1) Infrared Imaging: Detects organic protein residues.

(2) UV Fluorescence: Spots hidden blood components.

(3) High-Resolution Visible Light: Identifies physical scratches and surface flaws.

(4) Polarised Light: Catches oily film and chemical residues.

ii) Surface Discolouration Indicators: The solution should scan for visual flaws that reveal hidden damage, including:

(1) Corrosion: Early rust or oxidation spots that weaken the tool.

(2) Chemical Stains: Disinfectant marks that can hurt tool performance.

(3) Heat Damage: Burn marks or colour changes from high steriliser heat.

(4) Water Spots: Mineral deposits from hard water that can trap bacteria.

iii) Mechanical & Structural Defects: The solution should identify physical damage that compromises tool safety, such as:

(1) Chipped Edges: Damage on cutting or grasping parts (like scissors or forceps).

(2) Cracks: Hairline fractures in tool bodies, joints, or handles.

(3) Misalignment: Bent components or hinges that cause improper closure.

(4) Blunt Edges: Worn-out or dull cutting surfaces that affect precision.

(5) Loose Joints: Unstable or degraded joints and ratchet mechanisms.

(6) Surface Pitting: Tiny holes or worn material that weaken the tool's strength.

iv) Functional Degradation Indicators: The solution should detect signs of wear from repeated use, including:

(1) Deformation: Tools bent out of shape from mechanical stress or bad handling.

(2) Excessive Wear: General degradation that goes beyond acceptable limits.

v) Automated Quality Assurance: The solution should standardise pass/fail decisions across all shifts to eliminate subjective human variability. Items that fail predefined quality thresholds should be automatically flagged and rerouted for reprocessing as appropriate.

vi) Audit Readiness: The solution should generate automated compliance documentation on cleaning effectiveness, while using predictive analytics to identify instruments requiring preventive maintenance.

3. Final Packing Quality Check

i) Pre-Sterilisation Verification: The solution should perform image-based verification against reference databases to ensure instrument set completeness and identify any remaining structural defects before sterilisation. This includes verification of Chemical Indicators (CI) at packing.

ii) Infrastructure Integration: The solution should integrate seamlessly with existing CSSD systems (e.g. SSETs, T-Doc) to enable real-time workflow visibility and tracking.

iii) Traceability & Compliance: The solution should maintain end-to-end digital traceability to support regulatory compliance, enable precise batch tracking, and efficient recall management when required.  

For full details on the use cases, please click here.

• Please submit any questions or requests for clarification about the Challenge Statement in writing to healthx@synapxe.sg  by 23 Jul 2026, 5PM.
• Responses will be consolidated and published by 28 Jul 2026. Only published responses will be considered official clarifications to the Challenge Statement.
• For additional information, please visit our HealthX Call-for-Innovation FAQ section.