Chapter 3: Workflow and policy basics

3.1 One case, two worlds: glass vs digital workflow

What you need to know

A surgical pathology case follows the same biological path whether or not you are digital:
- specimen collection and transport
- accession and gross examination
- processing, embedding, cutting, staining, and coverslipping
- slide distribution, reporting, and archiving.
  Digital pathology adds steps; it does not remove the classical ones.
In a glass-only workflow, the glass slide is the only durable object for diagnosis:
- Old material is retrieved from the physical archive and placed in trays.
- Consultation, QA review, and tumour boards all depend on physically moving glass between benches, microscopes, and meeting rooms.
- Lost, broken, or misfiled slides are hard to recover; they represent real clinical risk.
In a glass + digital workflow, you add a parallel digital path:
- Before scanning, slides are labelled with standardised barcodes in scanner-friendly locations.
- A pre-scan QC step checks label legibility, coverslip quality, and tissue visibility.
- Slides are loaded into scanners; the system finds tissue, focuses, and acquires whole-slide images (WSIs) at a chosen magnification and microns-per-pixel.
- WSIs are transferred to an image server and linked to LIS cases, so the pathologist sees a digital worklist that mirrors their normal case list.
- WSIs are archived under a retention policy, just like glass slides, but can be retrieved instantly without physically pulling trays.
Early on, most labs run a hybrid phase:
- Glass is still the primary diagnostic medium.
- Digital images are used to:
  - compare current cases with old biopsies side-by-side
  - prepare for tumour boards by bookmarking key fields
  - obtain second opinions across sites without shipping glass.
- Over time, and after formal validation, selected case types may move to digital primary diagnosis with glass kept mainly as a safety backup.
Thinking like radiology helps:
- The tissue block is analogous to the patient.
- The glass slide is analogous to the original CT film era.
- The WSI plus image server is the PACS-like layer, which becomes the daily tool for viewing, QA, and MDT, while the physical material remains available in the background.
You should be able to draw both versions of the day for a typical case (for example, a colon cancer resection):
- glass-only: who handles the case, where delays occur, how old material is retrieved
- glass + digital: where scanning fits, what changes for histology staff, what changes for the reporting pathologist, and how MDT preparation improves.

Reference
Zarella MD, Bowman D, Aeffner F, et al. A practical guide to whole slide imaging: a white paper from the Digital Pathology Association. Arch Pathol Lab Med. 2019;143(2):222–234. doi:10.5858/arpa.2018-0343-RA.

3.2 Pre-analytic, analytic, and post-analytic: why the triad matters

What you need to know

Digital pathology guidance deliberately talks about pre-analytic, analytic, and post-analytic phases because:
- it matches long-standing laboratory medicine quality frameworks
- it forces you to look for failure modes in each phase rather than blaming “the scanner” for everything.
Pre-analytic phase (before the scanner):
- specimen identification and labelling
- gross examination, cassette labelling, and tissue sampling
- processing, embedding, cutting, staining, and coverslipping
- slide labelling and barcoding
- deciding which slides to scan and doing pre-scan QC (clean slide, intact coverslip, correct label).
  Pre-analytic errors include wrong patient, wrong site, mislabelling, poor tissue processing, folding, or staining failures. Digital systems cannot fix these; they only reproduce them at high resolution.
Analytic phase (scanner and viewer):
- loading slides into scanners and managing queues
- scanner focus, exposure, and image acquisition
- detection of gross scanning errors (missing tissue, out-of-focus scans, misreads of barcodes)
- transfer of WSIs to servers and linkage to LIS cases
- pathologists actually viewing and interpreting digital slides on calibrated workstations.
  Analytic failures include scanner malfunction, severe focus problems, truncated scans, or mismatched case/slide associations.
Post-analytic phase (after the diagnostic decision):
- entering, verifying, and releasing the report in the LIS/EHR
- archiving glass slides and blocks
- archiving WSIs according to retention policies, with regular checks that data can be restored
- using digital slides for MDTs, QA, EQA, education, and research.
  Post-analytic failures include reports not reaching the clinical team, misfiled slides, corrupted or unrecoverable digital archives, or untracked use of images outside governance frameworks.
For digital pathology, the triad is used to:
- design validation studies that include realistic pre- and post-analytic steps, not just scanner accuracy
- identify where to place quality indicators and audits (for example, pre-scan rejection rate, re-scan rate, archive restore success)
- structure incident reviews when something goes wrong (“Was this pre-analytic, analytic, or post-analytic?”).
You should be able to take a single case and label each step as pre-analytic, analytic, or post-analytic, and give at least one plausible error and safety check for each phase.

Reference
Cross S, Furness P, Igali L, Snead D, Treanor D. Best practice recommendations for implementing digital pathology. London: The Royal College of Pathologists; 2018. (G162).

3.3 The people making the workflow work: roles and responsibilities

What you need to know

Digital pathology is not “a scanner plus a viewer”; it is a coordinated activity of several roles. Services fail when these roles are vague or missing.
Typical key roles you should be able to name and describe:
- Digital pathology clinical lead (pathologist)
  - owns the clinical design of the digital service
  - leads or co-leads validation and signs off on clinical safety
  - helps define which cases are suitable for digital primary diagnosis and under what conditions.
- Histology / biomedical scientist lead
  - understands real bench workflow and constraints
  - decides how scanning fits alongside cutting and staining without disrupting turn-around times
  - designs pre-scan QC, rescanning routines, and test slide protocols.
- Scanner operators / designated technologists
  - load scanners, monitor queues and error messages
  - perform immediate quality checks (focus, completeness, barcode match)
  - are empowered to stop the line and revert to glass if scans or systems are unsafe.
- Reporting pathologists and trainees
  - use the viewer for sign-out, MDTs, QA, and teaching
  - are responsible for raising issues about image quality, linkage, or performance
  - participate in validation and ongoing QA (for example, periodic review sets).
- IT / LIS / imaging informatics support
  - maintain servers, storage, backups, and networks
  - manage interfaces between scanners, image servers, LIS, and viewers
  - handle user accounts, access control, and security patches.
- Quality / governance / information security
  - integrate digital slides into the existing quality management system
  - help design audits, risk registers, incident reviews, and SOPs
  - oversee privacy, retention, and use of images for teaching and research.
Good guidance emphasises that named individuals should be associated with each function, even if in smaller labs one person wears several hats. When something goes wrong, it must be clear whose responsibility it is to:
- investigate and document the issue
- decide whether to stop or restrict digital use temporarily
- implement and record corrective actions.
As a pathologist, you should be able to sketch a simple “who does what” table for your own lab that:
- covers pre-analytic, analytic, and post-analytic steps
- assigns a primary role to each step
- defines who is notified and who can decide to revert to glass when problems arise.

Reference
Cross S, Furness P, Igali L, Snead D, Treanor D. Best practice recommendations for implementing digital pathology. London: The Royal College of Pathologists; 2018. (G162).

3.4 Scanning policies: deciding what goes through the scanner

What you need to know

A scanning policy defines which slides are scanned, when, and why. It is a clinical and logistical decision, not a purely technical one.
There are several common starting patterns:
- Use-case based scanning
  - scan all slides for chosen use cases such as:
    - tumour board cases
    - external consultations
    - selected subspecialties (for example, dermatopathology, neuropathology)
    - cases used for teaching or resident exams.
- Case-type or specimen-based scanning
  - scan all malignant resections and their related biopsies
  - scan all small biopsies in certain organ systems
  - scan all cases with a specific diagnostic or prognostic impact.
- Location or pathway-based scanning
  - scan all cases from one hospital or subspecialty team as a pilot
  - use digital primary diagnosis only in selected clinical settings (for example, routine GI biopsies).
A sensible scanning policy must answer at least these questions:
1. Which cases or slides are in scope?
  - by diagnosis, specimen type, use case, or location.
2. When are they scanned?
  - same day as staining, overnight, or on demand.
3. Who decides and who checks?
  - how technologists know what belongs in the scanner racks and who performs pre-scan QC.
4. What happens when scanning fails or is delayed?
  - criteria for reverting to glass, prioritising urgent cases, and communicating delays.
5. How does this interact with validation and training?
  - pathologists should not be forced into digital primary diagnosis for case types that have not been validated.
Over time, many services evolve from partial scanning to broader coverage as:
- confidence grows through validation and everyday use
- infrastructure (storage, network) proves adequate
- clinicians and administrators see clear benefits for MDTs, QA, and service resilience.
As a pathologist-in-training or digital lead, you should be able to:
- draft a first scanning policy for your lab, based on realistic capacity
- explain the clinical rationale (why these cases first, and what problem you are solving)
- identify what data and metrics (for example, scan failure rate, time from staining to available WSI) you would monitor to refine the policy.

3.5 Failure points and safety nets in the digital pathology workflow

What you need to know

Digital pathology introduces new failure modes but also new opportunities for safety checks. You must think explicitly about both.
Typical pre-analytic weak spots:
- wrong patient, wrong site, or wrong side at accession or grossing
- mislabelled cassettes or slides; barcodes not matching LIS data
- poor tissue processing leading to incomplete sections, chatter, or severe folds
- staining failures that make tissue unreadable; labels obscuring tissue.
  Safety nets include:
- robust barcoding with electronic order linkage
- standardised slide labelling layouts that keep labels away from tissue
- defined acceptance criteria for slides to enter the scanning pipeline (clean, dry, intact, correctly labelled)
- training and competency checks for technologists.
Typical analytic weak spots:
- scanner misfocus (especially at tissue edges or thick sections)
- incomplete scans (missing tiles, truncated tissue)
- misregistered slides or mislinked WSIs to LIS cases
- server, storage, or network outages that prevent viewing.
  Safety nets include:
- routine test slides and calibration procedures for scanners
- clear criteria for when to re-scan and who decides
- QC steps where operators briefly examine images before releasing them to reporting
- monitoring tools and alerts for storage capacity, network latency, and service status
- documented downtime procedures for reversion to glass.
Typical post-analytic weak spots:
- report entered on the wrong patient or case when multiple viewers or windows are open
- digital archives that silently fail to back up or cannot be restored in a crisis
- uncontrolled reuse of images in teaching, presentations, or research without approvals.
  Safety nets include:
- workflow designs that minimise the chance of mismatched viewer/LIS cases (for example, integrated viewers)
- periodic restore drills from backup to prove that archives are usable
- governance policies and training for secondary use of images, including consent and anonymisation requirements.
High-quality guidelines treat digital pathology as a system that:
- must be validated locally in the context of your own pre-analytic, analytic, and post-analytic processes
- should be subject to ongoing QA and incident review, not a “once and done” project
- requires transparent documentation so that external reviewers can understand how you manage risks.
As a learner, you should be able to:
- list at least five realistic failure points in your own lab’s digital workflow
- map each one to one or more safety nets
- explain how these safety nets are monitored over time (audits, KPIs, incident logs).

Reference
Evans AJ, Brown RW, Bui MM, et al. Validating whole slide imaging systems for diagnostic purposes in pathology: guideline update from the College of American Pathologists. Arch Pathol Lab Med. 2022;146(4):440–450. doi:10.5858/arpa.2020-0723-CP.

--- title: "Chapter 3: Workflow and policy basics" format: html: toc: true toc-depth: 3 code-fold: false df-print: paged smooth-scroll: true execute: echo: true warning: false message: false --- # Chapter 3: Workflow and policy basics ## 3.1 One case, two worlds: glass vs digital workflow **What you need to know** - A surgical pathology case follows the **same biological path** whether or not you are digital: - specimen collection and transport - accession and gross examination - processing, embedding, cutting, staining, and coverslipping - slide distribution, reporting, and archiving. Digital pathology **adds** steps; it does not remove the classical ones. - In a **glass-only workflow**, the glass slide is the only durable object for diagnosis: - Old material is retrieved from the physical archive and placed in trays. - Consultation, QA review, and tumour boards all depend on **physically moving glass** between benches, microscopes, and meeting rooms. - Lost, broken, or misfiled slides are hard to recover; they represent real clinical risk. - In a **glass + digital workflow**, you add a parallel digital path: - Before scanning, slides are labelled with **standardised barcodes** in scanner-friendly locations. - A pre-scan QC step checks label legibility, coverslip quality, and tissue visibility. - Slides are loaded into scanners; the system finds tissue, focuses, and acquires whole-slide images (WSIs) at a chosen magnification and microns-per-pixel. - WSIs are transferred to an image server and **linked to LIS cases**, so the pathologist sees a digital worklist that mirrors their normal case list. - WSIs are archived under a retention policy, just like glass slides, but can be retrieved instantly without physically pulling trays. - Early on, most labs run a **hybrid phase**: - **Glass is still the primary** diagnostic medium. - Digital images are used to: - compare current cases with old biopsies side-by-side - prepare for tumour boards by bookmarking key fields - obtain second opinions across sites without shipping glass. - Over time, and after formal validation, selected case types may move to **digital primary diagnosis** with glass kept mainly as a safety backup. - Thinking like radiology helps: - The tissue block is analogous to the patient. - The glass slide is analogous to the original CT film era. - The WSI plus image server is the **PACS-like layer**, which becomes the daily tool for viewing, QA, and MDT, while the physical material remains available in the background. - You should be able to **draw both versions of the day** for a typical case (for example, a colon cancer resection): - glass-only: who handles the case, where delays occur, how old material is retrieved - glass + digital: where scanning fits, what changes for histology staff, what changes for the reporting pathologist, and how MDT preparation improves. **Reference** Zarella MD, Bowman D, Aeffner F, et al. A practical guide to whole slide imaging: a white paper from the Digital Pathology Association. *Arch Pathol Lab Med.* 2019;143(2):222–234. doi:10.5858/arpa.2018-0343-RA. --- ## 3.2 Pre-analytic, analytic, and post-analytic: why the triad matters **What you need to know** - Digital pathology guidance deliberately talks about **pre-analytic, analytic, and post-analytic** phases because: - it matches long-standing laboratory medicine quality frameworks - it forces you to look for failure modes in each phase rather than blaming “the scanner” for everything. - **Pre-analytic phase** (before the scanner): - specimen identification and labelling - gross examination, cassette labelling, and tissue sampling - processing, embedding, cutting, staining, and coverslipping - slide labelling and barcoding - deciding **which slides to scan** and doing pre-scan QC (clean slide, intact coverslip, correct label). Pre-analytic errors include wrong patient, wrong site, mislabelling, poor tissue processing, folding, or staining failures. Digital systems **cannot fix** these; they only reproduce them at high resolution. - **Analytic phase** (scanner and viewer): - loading slides into scanners and managing queues - scanner focus, exposure, and image acquisition - detection of gross scanning errors (missing tissue, out-of-focus scans, misreads of barcodes) - transfer of WSIs to servers and **linkage to LIS cases** - pathologists actually viewing and interpreting digital slides on calibrated workstations. Analytic failures include scanner malfunction, severe focus problems, truncated scans, or mismatched case/slide associations. - **Post-analytic phase** (after the diagnostic decision): - entering, verifying, and releasing the report in the LIS/EHR - archiving glass slides and blocks - archiving WSIs according to retention policies, with regular checks that data can be restored - using digital slides for MDTs, QA, EQA, education, and research. Post-analytic failures include reports not reaching the clinical team, misfiled slides, corrupted or unrecoverable digital archives, or untracked use of images outside governance frameworks. - For digital pathology, the triad is used to: - design **validation studies** that include realistic pre- and post-analytic steps, not just scanner accuracy - identify where to place **quality indicators** and audits (for example, pre-scan rejection rate, re-scan rate, archive restore success) - structure **incident reviews** when something goes wrong (“Was this pre-analytic, analytic, or post-analytic?”). - You should be able to take a single case and label each step as pre-analytic, analytic, or post-analytic, and give at least one **plausible error** and **safety check** for each phase. **Reference** Cross S, Furness P, Igali L, Snead D, Treanor D. Best practice recommendations for implementing digital pathology. London: The Royal College of Pathologists; 2018. (G162). --- ## 3.3 The people making the workflow work: roles and responsibilities **What you need to know** - Digital pathology is **not** “a scanner plus a viewer”; it is a coordinated activity of several roles. Services fail when these roles are vague or missing. - Typical key roles you should be able to name and describe: - **Digital pathology clinical lead (pathologist)** - owns the clinical design of the digital service - leads or co-leads validation and signs off on clinical safety - helps define which cases are suitable for digital primary diagnosis and under what conditions. - **Histology / biomedical scientist lead** - understands real bench workflow and constraints - decides how scanning fits alongside cutting and staining without disrupting turn-around times - designs pre-scan QC, rescanning routines, and test slide protocols. - **Scanner operators / designated technologists** - load scanners, monitor queues and error messages - perform immediate quality checks (focus, completeness, barcode match) - are empowered to **stop the line** and revert to glass if scans or systems are unsafe. - **Reporting pathologists and trainees** - use the viewer for sign-out, MDTs, QA, and teaching - are responsible for raising issues about image quality, linkage, or performance - participate in validation and ongoing QA (for example, periodic review sets). - **IT / LIS / imaging informatics support** - maintain servers, storage, backups, and networks - manage interfaces between scanners, image servers, LIS, and viewers - handle user accounts, access control, and security patches. - **Quality / governance / information security** - integrate digital slides into the existing quality management system - help design audits, risk registers, incident reviews, and SOPs - oversee privacy, retention, and use of images for teaching and research. - Good guidance emphasises that **named individuals** should be associated with each function, even if in smaller labs one person wears several hats. When something goes wrong, it must be clear whose responsibility it is to: - investigate and document the issue - decide whether to stop or restrict digital use temporarily - implement and record corrective actions. - As a pathologist, you should be able to sketch a simple **“who does what” table** for your own lab that: - covers pre-analytic, analytic, and post-analytic steps - assigns a primary role to each step - defines who is notified and who can decide to revert to glass when problems arise. **Reference** Cross S, Furness P, Igali L, Snead D, Treanor D. Best practice recommendations for implementing digital pathology. London: The Royal College of Pathologists; 2018. (G162). --- ## 3.4 Scanning policies: deciding what goes through the scanner **What you need to know** - A **scanning policy** defines which slides are scanned, when, and why. It is a clinical and logistical decision, not a purely technical one. - There are several common starting patterns: - **Use-case based scanning** - scan all slides for chosen use cases such as: - tumour board cases - external consultations - selected subspecialties (for example, dermatopathology, neuropathology) - cases used for teaching or resident exams. - **Case-type or specimen-based scanning** - scan all malignant resections and their related biopsies - scan all small biopsies in certain organ systems - scan all cases with a specific diagnostic or prognostic impact. - **Location or pathway-based scanning** - scan all cases from one hospital or subspecialty team as a pilot - use digital primary diagnosis only in selected clinical settings (for example, routine GI biopsies). - A sensible scanning policy must answer at least these questions: 1. **Which cases or slides are in scope?** - by diagnosis, specimen type, use case, or location. 2. **When are they scanned?** - same day as staining, overnight, or on demand. 3. **Who decides and who checks?** - how technologists know what belongs in the scanner racks and who performs pre-scan QC. 4. **What happens when scanning fails or is delayed?** - criteria for reverting to glass, prioritising urgent cases, and communicating delays. 5. **How does this interact with validation and training?** - pathologists should not be forced into digital primary diagnosis for case types that have not been validated. - Over time, many services evolve from **partial scanning** to broader coverage as: - confidence grows through validation and everyday use - infrastructure (storage, network) proves adequate - clinicians and administrators see clear benefits for MDTs, QA, and service resilience. - As a pathologist-in-training or digital lead, you should be able to: - draft a **first scanning policy** for your lab, based on realistic capacity - explain the clinical rationale (why these cases first, and what problem you are solving) - identify what data and metrics (for example, scan failure rate, time from staining to available WSI) you would monitor to refine the policy. **Reference** Zarella MD, Bowman D, Aeffner F, et al. A practical guide to whole slide imaging: a white paper from the Digital Pathology Association. *Arch Pathol Lab Med.* 2019;143(2):222–234. doi:10.5858/arpa.2018-0343-RA. --- ## 3.5 Failure points and safety nets in the digital pathology workflow **What you need to know** - Digital pathology introduces **new failure modes** but also new opportunities for safety checks. You must think explicitly about both. - Typical **pre-analytic weak spots**: - wrong patient, wrong site, or wrong side at accession or grossing - mislabelled cassettes or slides; barcodes not matching LIS data - poor tissue processing leading to incomplete sections, chatter, or severe folds - staining failures that make tissue unreadable; labels obscuring tissue. Safety nets include: - robust barcoding with electronic order linkage - standardised slide labelling layouts that keep labels away from tissue - defined acceptance criteria for slides to enter the scanning pipeline (clean, dry, intact, correctly labelled) - training and competency checks for technologists. - Typical **analytic weak spots**: - scanner misfocus (especially at tissue edges or thick sections) - incomplete scans (missing tiles, truncated tissue) - misregistered slides or mislinked WSIs to LIS cases - server, storage, or network outages that prevent viewing. Safety nets include: - routine **test slides** and calibration procedures for scanners - clear criteria for when to **re-scan** and who decides - QC steps where operators briefly examine images before releasing them to reporting - monitoring tools and alerts for storage capacity, network latency, and service status - documented **downtime procedures** for reversion to glass. - Typical **post-analytic weak spots**: - report entered on the wrong patient or case when multiple viewers or windows are open - digital archives that silently fail to back up or cannot be restored in a crisis - uncontrolled reuse of images in teaching, presentations, or research without approvals. Safety nets include: - workflow designs that minimise the chance of mismatched viewer/LIS cases (for example, integrated viewers) - periodic **restore drills** from backup to prove that archives are usable - governance policies and training for secondary use of images, including consent and anonymisation requirements. - High-quality guidelines treat digital pathology as a system that: - must be **validated locally** in the context of your own pre-analytic, analytic, and post-analytic processes - should be subject to ongoing QA and incident review, not a “once and done” project - requires transparent documentation so that external reviewers can understand how you manage risks. - As a learner, you should be able to: - list at least **five realistic failure points** in your own lab’s digital workflow - map each one to one or more **safety nets** - explain how these safety nets are monitored over time (audits, KPIs, incident logs). **Reference** Evans AJ, Brown RW, Bui MM, et al. Validating whole slide imaging systems for diagnostic purposes in pathology: guideline update from the College of American Pathologists. *Arch Pathol Lab Med.* 2022;146(4):440–450. doi:10.5858/arpa.2020-0723-CP.