AI Detection of Equine colic?

Hello, Summarians!

AI seems to be showing up in a variety of ways. We will have to look at the uses critically to see if it lives up to the hype.

Please give me feedback on ways I can make it more useful to you.

If it is valuable to you please consider sending it on to a friend 😄 

Bleeding Risk with NSAID Eye Drops

The study aimed to investigate the incidence of gastrointestinal (GI) bleeding in dogs treated with topical ophthalmic nonsteroidal anti-inflammatory drugs (NSAIDs) and compare it with the incidence in dogs treated with systemic NSAIDs or glucocorticoids, which are known to increase the risk of GI bleeding. The retrospective analysis found that dogs receiving ophthalmic NSAIDs had a similar frequency of GI bleeding as those receiving systemic NSAIDs or glucocorticoids alone, suggesting that ophthalmic NSAIDs might independently pose a risk for GI bleeding. This risk appeared consistent even when excluding dogs with concurrent risk factors for bleeding or those also receiving systemic NSAIDs or glucocorticoids, indicating that co-administration did not significantly increase the bleeding risk beyond that of systemic administration alone. Notably, severe GI bleeding was only observed in dogs treated with ketorolac, the ophthalmic NSAID delivered at the highest concentration, suggesting a potential for greater systemic absorption and associated risk. The study highlights limitations such as potential underestimation of GI adverse effects and inconsistent monitoring, pointing to the need for prospective research to better understand the systemic absorption and GI risk of ophthalmic NSAIDs in dogs. 

 Van Vertloo LR, Terhaar HM, Viall AK, Allbaugh RA. Retrospective evaluation of the incidence of gastrointestinal bleeding in dogs receiving ophthalmic nonsteroidal anti-inflammatory drugs. Vet Ophthalmol. 2023; 26: 560-564. doi:10.1111/vop.13145 

Bottom line — May be similar to oral NSAID use.

Stelfonta at 1 year.

The study conducted by Pamela Jones, DVM, DACVIM (Oncology), DACVR (Radiation oncology), and colleagues aimed to evaluate the clinical use and efficacy of Stelfonta® (tigilanol tiglate), an FDA-approved intratumoral injection for treating canine mast cell tumours (MCTs). The research encompassed data from 170 patients with 173 MCTs, primarily located on the limbs and diagnosed through cytology or histopathology. Notably, 68% of the treated tumours achieved complete remission, with the median time to best response being 28 days. The study also highlighted that the median wound size post-treatment was 4.17 cm², with wounds typically healing within 30 days. Adjunct treatments included prednisone, H2 blockers, H1 blockers, pain medications, among others. The overall findings indicated that Stelfonta is well-tolerated and effective, with 87% of participating oncologists willing to recommend its use for managing canine MCTs. 

(2024), ESVONC abstracts. Vet Comp Oncol, 22: 3-19. https://doi.org/10.1111/vco.12951 

Bottom line — Well tolerated and effective.

Wearables and AI to Assess Colic Risk

The study aimed to develop a software algorithm capable of aiding in the detection of colic signs and levels of pain in horses, utilizing advancements in wearables and artificial intelligence. Through in vivo experiments on eight experimental mares induced with transient colic, the study collected accelerometric data and videos to observe and classify the horses' behavior into normal and 10 pain-related behaviors. Veterinarians assigned pain scores to separate colic episodes into none, level 1, or level 2 based on severity, which served as a ground truth for validating the software model's accuracy. 

The developed model demonstrated a high accuracy rate, with 91.2% for detecting colic presence and 93.8% for differentiating between level 1 and level 2 colic severity. The system was capable of accurately classifying pain-related behaviors, indicating its potential as a tool for early detection of colic in horses and assessing the severity of their condition. 

The study's main limitation was the inclusion of a limited number of horses with severe pain-related behaviors, which could affect the accuracy of categorizing colic severity. Nonetheless, the conclusions highlighted the system's uniqueness and innovation in early colic detection and severity differentiation, presenting a significant advancement in veterinary diagnostic capabilities through the use of artificial intelligence and behavioral monitoring. 

Eerdekens A, Papas M, Damiaans B, Martens L, Govaere J, Joseph W, Deruyck M. Automatic early detection of induced colic in horses using accelerometer devices. Equine Vet J. 2024 Feb 6. doi: 10.1111/evj.14069. PMID: 38318654 

Bottom line — Shows promise in early studies.

Just putting things in perspective …