Swallow Study (VFSS)

What it does

Swallow Study reads a videofluoroscopic swallow study (VFSS), also called a modified barium swallow, run by run. Each cine run is one bolus; the run follows the bolus, the hyoid bone and the upper oesophageal sphincter (UES) through every frame, and marks when material enters the laryngeal vestibule or passes below the vocal folds.

For each bolus it returns the penetration and aspiration events with their frame ranges and a Penetration–Aspiration Scale score, then the worst score of the study. Dynamic measurements (hyoid excursion, UES opening, transit times and residue left in the pharynx) describe how the swallow is impaired, not only its outcome.

Intended use

Decision support for the clinician who reviews the study, usually a speech-language pathologist with a radiologist. Reading a VFSS frame by frame is slow, and brief aspiration events are easy to miss at normal playback. The run points to the frames that matter and fills in the scores that guide diet and therapy.

Who it is for

Speech-language pathologists, radiologists, ENT surgeons, stroke and rehabilitation physicians, and teams caring for head-and-neck cancer and frail elderly patients. Developers can attach the run to a fluoroscopy archive or a dysphagia clinic record.

Accepted input

• DICOM X-Ray Radiofluoroscopic cine runs, including Enhanced XRF multi-frame objects.
• One run per bolus, lateral projection, with optional AP runs; a study usually holds several boluses of different consistencies.
• Typical runs last 10 to 60 seconds.

Requirements

Requirement	Why
Frame rate of 15 fps or more	Penetration and aspiration can last only a few frames
CineRate or FrameTime	Transit times and event timing are computed from the frame clock
Lateral projection	The laryngeal vestibule, vocal folds and UES are judged on the lateral view
Bolus consistency per run	Scores are reported per consistency; IDDSI levels are the expected labels
Collimation including the oral cavity to the cervical oesophagus	Oral transit, pharyngeal residue and UES opening need the whole pathway

Low frame rates, tight collimation and swallows split across runs are flagged in the result.

Optional context

The indication (stroke, head-and-neck cancer, progressive neurological disease), current diet and the IDDSI level of each bolus can be given in clinical_context.

The result

A JSON result with a bolus[] list (consistency, PAS score, frame range) and the two sections below; key frames and segmentation masks come with every event.

Section	Content
Penetration and aspiration	Events per bolus with frame ranges, PAS score 1–8 per bolus and the worst PAS of the study
Swallow kinematics	Bolus, hyoid and UES segmentation; hyoid excursion, UES opening, oral and pharyngeal transit times and residue
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

Standards

• Penetration–Aspiration Scale (Rosenbek 1996), scored 1 to 8 per bolus.
• MBSImP components for the kinematic measurements.
• IDDSI framework for bolus consistency levels.
• DICOM: TID 1500 Measurement Report for scores and timings; DICOM SEG for bolus, hyoid and UES masks; Encapsulated PDF.

Reading the scores

PAS 1 means no material enters the airway. Scores 2 to 5 describe penetration into the laryngeal vestibule and 6 to 8 aspiration below the vocal folds; within each group the score depends on the depth reached and on whether the material is ejected. PAS 8 is silent aspiration: material below the vocal folds with no effort to eject it. The study-level score is the worst bolus.