Artificial intelligence in veterinary diagnostic imaging: A literature review.

Artificial intelligence in veterinary medicine is an emerging field. Machine learning, a subfield of artificial intelligence, allows computer programs to analyze large imaging datasets and learn to perform tasks relevant to veterinary diagnostic imaging. This review summarizes the small, yet growing body of artificial intelligence literature in veterinary imaging, provides necessary background to understand these papers, and provides author commentary on the state of the field. To date, less than 40 peer-reviewed publications have utilized machine learning to perform imaging-associated tasks across multiple anatomic regions in veterinary clinical and biomedical research. Major challenges in this field include collection and cleaning of sufficient image data, selection of high-quality ground truth labels, formation of relationships between veterinary and machine learning professionals, and closure of the gap between academic uses of artificial intelligence and currently available commercial products. Further development of artificial intelligence has the potential to help meet the growing need for radiological services through applications in workflow, quality control, and image interpretation for both general practitioners and radiologists.

Clinical and diagnostic evaluation of intraocular expulsion of a corneal epithelial inclusion cyst in a dog.

CASE DESCRIPTION: An 8-year-old spayed female mixed-breed dog was presented for evaluation of an axial, raised, vascularized, lightly pigmented 3 mm diameter corneal lesion affecting the left eye (OS) that was present since adoption three years prior to presentation. The presumed cyst ruptured into the anterior chamber 5 months following initial presentation with progressive extrusion of intralesional contents into the anterior chamber. High-frequency ultrasound and in vivo confocal microscopy revealed minimal normal stroma posterior to the corneal lesion. Due to the lack of stroma, a deep anterior lamellar keratoplasty (DALK) was attempted using the viscodissection technique. While separating the stroma from Descemet's membrane, a tear in the membrane was observed, and the procedure was converted to a penetrating keratoplasty (PK). The mass was excised en bloc, and a frozen corneal allograft and conjunctival pedicle graft were utilized to restore corneal thickness. The extruded material was irrigated out of the anterior chamber and submitted for culture and cytology. There was no growth on aerobic, anaerobic, or fungal cultures, and cytology revealed mixed neutrophilic and macrophagic inflammation with keratinizing squamous epithelium. Histopathology identified the mass to be a corneal epithelial inclusion cyst lined with well-differentiated stratified squamous epithelium. The dog is doing well 10 months post-operatively with no signs of recurrence along with good comfort and vision.

Evaluation of a 3-dimensional ultrasound device for noninvasive measurement of urinary bladder volume in dogs.

BACKGROUND: The BladderScan Prime Plus (BPP; Verathon, Bothell, Washington) is an application-specific, three-dimensional ultrasound device used for human, point-of-care volumetry of the urinary bladder. OBJECTIVE: To estimate the BPP's accuracy, repeatability, and optimized settings for assessing urinary bladder volumes in dogs, a variable utilized in assessing micturition disorders. ANIMALS: Twenty-four, client-owned, healthy, male dogs presenting for routine examination. METHODS: Prospective examinations were conducted by an experienced ultrasonographer and a novice, selecting the BPP's "man" or "child" setting, and were compared to urine volume obtained by catheterization. RESULTS: Mean urine volume significantly varied by operator (P = .05), device setting (P < .001), and weight (P = .01); the "man" setting produced mean volumes nearer to catheterized volumes. The mean difference between BPP's "man" setting and catheterized volume was 0.88 mL, with maximal positive and negative disagreement of +23.2 mL to -55.3 mL (SD 19.0). Percent disagreement between BPP and catheterized volumes demonstrated a mean of -4.5%, with maximal positive and negative disagreement of +58.1% to -74.1% (SD 34.9). The experienced operator recorded volumes significantly (P = .05) higher than the novice, with difference in means of 3.2 mL. In dogs weighing >5.5 kg (n = 18/24), mean difference between BPP's "man" setting and catheterized measurements, regardless of operator, was not significant. CONCLUSIONS: Although small magnitude interuser variability is present in BPP examinations, the device provides accurate, though imprecise quantification of bladder volume in canids weighing >5.5 kg.