Laryngoscopy and Stroboscopy

What it does

Laryngoscopy and Stroboscopy reads recorded examinations of the larynx: flexible or rigid laryngoscopy, and stroboscopic or high-speed recordings of the vibrating vocal folds. These are the examinations behind the work-up of hoarseness and of suspected laryngeal cancer.

The run returns two kinds of result. On laryngoscopy frames it flags vocal-fold lesions: nodules, polyps, cysts, leukoplakia and lesions suspicious for malignancy, each with a key frame. On stroboscopy or high-speed video it segments the glottal area, the opening between the vocal folds, in every frame, and turns the area over time into a glottal area waveform from which glottic function measurements are computed.

Intended use

Decision support after the examination: structured lesion flags for the ENT report and a quantified view of vocal-fold vibration for voice assessment and follow-up after therapy or surgery. Every lesion flag links to its key frame, and recordings from repeat visits give waveforms that can be set side by side.

Who it is for

Otolaryngologists, laryngologists and phoniatricians, speech and language therapists who work with stroboscopy, and voice clinics that track patients over time. Researchers and developers use the glottis masks and waveforms for voice analysis at scale.

Accepted input

• Laryngoscopy video or stills: flexible or rigid examinations as MP4, MOV or MKV, DICOM Video Endoscopic Image, or JPEG/PNG and DICOM VL Endoscopic Image stills.
• Stroboscopy and high-speed video: recordings of phonation, used for the glottis segmentation and waveform.

Requirements

Requirement	Why
Vocal folds fully in view	Lesion flags and the glottal area need both folds unobstructed
Stroboscopy or high-speed recording for function	Vocal-fold vibration is too fast to follow in plain video
True frame rate preserved in the file	Waveform timing is computed from it
Laryngoscopy frames with good light and focus	Leukoplakia and small lesions are judged on surface detail

Optional context

Symptoms and their duration, voice use, smoking history and prior laryngeal surgery or therapy can be passed in clinical_context; they shape the guidance and the draft report.

The result

Section	Content
Laryngeal lesions	Vocal-fold nodules, polyps, cysts, leukoplakia and suspected malignancy, with key frames
Glottis segmentation	Glottal-area mask per frame, glottal area waveform and glottic function measurements
Draft report	Opt-in (options.report): an English narrative that interprets the result for the reader (findings, impression, limitations); every number is checked against the findings

Masks, waveforms and lesion flags come back together in one JSON result, with each value linked to the frames it was measured on.

Standards

• DICOM: glottis masks as Segmentation, key frames as VL Endoscopic Image with a Key Object Selection, annotated clips as Video Endoscopic Image, measurements and findings in SR TID 1500, an Encapsulated PDF report.
• Interoperability: FHIR R4 Procedure, DiagnosticReport and Observation; SNOMED CT codes for finding and anatomic site.