fix(defaultRouteInit): defer prior fetches and guard study browser thumbnails

[nutzlastfan]

Context

This ports a downstream loading fix that improves the first hanging-protocol application while still pulling prior studies in the background.

The downstream issue had two parts:

• active and prior study retrievals were mixed too early during route initialization, which made the first HP run less predictable
• the Study Browser could throw when a display set had no usable thumbnail image ids

Changes

• fetch and store patient studies up front, but keep the initial metadata retrieval focused on the active study set
• start prior-study series retrieval only after the first required series have loaded and the initial hanging protocol has been applied
• upsert study metadata in DicomMetadataStore instead of assuming a fresh entry every time
• guard Study Browser thumbnail selection for display sets without image ids or dynamic timepoints

Result

• more stable initial study loading
• priors still get prefetched, but they no longer interfere with the first layout decision
• no Study Browser crash when SR/SEG/RT or other non-thumbnail-ready display sets are present

Notes

• Opened as a draft because defaultRouteInit has been evolving quickly on upstream and I expect architecture-level feedback on the prefetch timing.

Greptile Summary

This PR restructures defaultRouteInit to separate active-study metadata retrieval from patient-history/prior-study fetching, deferring prior retrieves until after the first HP application, and adds defensive null guards to getImageIdForThumbnail for display sets that have no image IDs (SR/SEG/RT). The Study Browser guard is clean. The defaultRouteInit refactor is directionally correct but has two P2 concerns around prior-study filter leakage and missing cleanup on navigation.

Confidence Score: 4/5

Safe to merge with awareness of two P2 concerns; active-study loading and initial HP application are not regressed.

No P0/P1 findings. Two P2 issues: active-study seriesInstanceUID filter leaking into prior-study retrieval (silent suppression of prior series when a series filter is set), and a fire-and-forget startPriorFetches that lacks a cancellation path on route change.

platform/app/src/routes/Mode/defaultRouteInit.ts — specifically startPriorFetches and the filters forwarded to prior retrieves.

Important Files Changed

Filename	Overview
platform/app/src/routes/Mode/defaultRouteInit.ts	Introduces fetchAndStorePatientStudies, upsertStudyMetadata, collectSeriesPromises, and startPriorFetches to defer prior-study fetching after initial HP application; fire-and-forget prior fetches lack cancellation and the active-study filters are forwarded to prior-study retrieval without stripping seriesInstanceUID.
extensions/default/src/Panels/StudyBrowser/PanelStudyBrowser.tsx	Adds null/empty guards for imageIds, timePoints, and middleTimePointImageIds in getImageIdForThumbnail; both call sites already handle an undefined return value, so the guards are safe and correct.

Prompt To Fix All With AI

Fix the following 2 code review issues. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 2
platform/app/src/routes/Mode/defaultRouteInit.ts:267-274
**`filters` applied to prior-study retrieval may suppress all prior series**

`startPriorFetches` passes the same `filters` object as the active-study retrieval. If `filters.seriesInstanceUID` is set (e.g., `?SeriesInstanceUID=1.2.3`), the data source will only try to retrieve that specific series UID from every prior study — UIDs that won't exist in those studies — so each prior study returns zero series. The `collectSeriesPromises` loop then finds nothing, `requiredSeriesPromises` stays empty, and the prior studies silently load no display sets.

Since `seriesInstanceUID` filtering is meaningful only for the active study, passing a separate `filters` object (or stripping `seriesInstanceUID`) for prior retrieves would avoid this silent suppression.

```typescript
const priorFilters = filters
  ? { ...filters, seriesInstanceUID: undefined }
  : filters;

const priorRetrieves = priorStudyUIDs.map(StudyInstanceUID =>
  dataSource.retrieve.series.metadata({
    StudyInstanceUID,
    filters: priorFilters,
    returnPromises: true,
    sortCriteria: customizationService.getCustomization('sortingCriteria'),
  })
);
```

### Issue 2 of 2
platform/app/src/routes/Mode/defaultRouteInit.ts:303-305
**`startPriorFetches` has no cancellation / cleanup on route change**

`startPriorFetches` is fired as a fire-and-forget and is never added to `unsubscriptions`. If the user navigates to a new study before it completes, it continues running in the old closure and will call `applyHangingProtocol()` (line 287) against the shared `hangingProtocolService`/`displaySetService` singletons, potentially re-applying the prior route's HP ID over the new study's layout, and may add stale study metadata to `DicomMetadataStore`.

A cancellation token or an `AbortController`-style flag checked before each `applyHangingProtocol()` call and before each metadata upsert would prevent this cross-route interference.

_{Reviews (4): Last reviewed commit: "fix(app): limit URL display-set loading ..." | Re-trigger Greptile}

[netlify]

✅ Deploy Preview for ohif-dev ready!

Name	Link
🔨 Latest commit	65c763d
🔍 Latest deploy log	https://app.netlify.com/projects/ohif-dev/deploys/69f363a418637900086eca37
😎 Deploy Preview	https://deploy-preview-5971--ohif-dev.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.
🤖 Make changes	Run an agent on this branch

---

To edit notification comments on pull requests, go to your Netlify project configuration.