New World screwworm (Cochliomyia hominivorax) is now one of the most urgent biosecurity threats facing North American cattle. The first operational challenge is simple but enormous: find a suspicious wound quickly, identify the animal, preserve the evidence, and move that case to inspection before routine movement gives the parasite more time.
Livestock Technologies is developing a Nose ID screwworm-screening system for cattle chutes. The design combines identity evidence with fixed color and thermal cameras so feedlots, livestock markets, border facilities, and research sites can create a repeatable whole-animal screening record while cattle are already moving through a controlled handling point.
Why cattle operations need scalable screwworm screening
USDA's current response site now provides a dashboard of confirmed U.S. animal and wild-fly detections and directs anyone who sees suspicious wounds, maggots, or infestations to report them immediately. USGS describes early detection and rapid response as critical and is separately validating a rapid field DNA test because visual similarities between fly species make expert confirmation necessary.
Manual head-to-tail inspection remains essential, especially for hidden areas a fixed camera cannot see. But asking people to perform the same detailed inspection across thousands of cattle creates a scale problem. Automated chute screening is meant to narrow that workload: consistently capture every animal, surface the passages that deserve attention, and retain the original evidence for review.
Why federal research is moving toward AI wound surveillance
USDA's published spending plan includes Texas A&M AgriLife work on AI-enabled, multimodal early wound surveillance for beef and dairy operations. A separate AgriLife project focuses on non-invasive sensing of screwworm larvae. Other selected teams are working on traps, field diagnostics, therapeutics, surveillance, and response tools.
That research direction matches the practical problem our Nose ID chute is built to address: faster, scalable surveillance around wounds—the place where New World screwworm begins. USDA has not endorsed Livestock Technologies, and the award is not funding our company; it is independent evidence that multimodal wound surveillance is now a serious national research priority.
The complete project names are available in the official APHIS recommended spending plan.
Why an AI cattle chute is the right screening point
A chute gives computer vision something the open pen rarely does: a known path, a constrained distance, repeatable lighting, and several seconds to collect multiple views. It is a practical place to build an auditable inspection record while cattle are already being handled.
But the sensor geometry matters. Nose ID focuses on face and muzzle imagery. Screwworm-related wounds can occur elsewhere on the animal, so a serious screening workflow needs additional side and oblique body views. Color cameras can surface visible wound-like regions, drainage, or blood. Thermal cameras can surface localized temperature asymmetry. Neither signal is specific enough to name the cause.
How the Nose ID AI screwworm detection system works
We are adding a multimodal screwworm wound-risk workflow to the Nose ID research roadmap. It turns each chute passage into an identity-linked inspection opportunity: acquire consistent views, reject poor captures, screen for visual and thermal risk signals, and move questionable evidence to the right person faster.
- 01
Identify the chute passage
Start with the loading record, tag, or manifest. Nose ID imagery builds supporting animal-identity evidence as biometric matching is validated.
- 02
Capture synchronized body views
Fixed RGB and thermal cameras collect side and oblique views. A front-facing Nose ID camera alone cannot inspect the full hide.
- 03
Screen for visible risk signals
Quality gates reject poor views before models surface wound-like regions, surface changes, fly activity, or thermal asymmetry.
- 04
Escalate the evidence
Send original frames, location, confidence, and capture gaps to an operator or veterinarian for immediate inspection and reporting decisions.
The system may help flag
- Visible wound-like regions or fresh surface changes
- Localized thermal asymmetry for human review
- Missing, blurred, or obstructed body views
- A time-stamped evidence packet tied to the chute passage
The system will not claim
- That a wound contains New World screwworm
- That an unflagged animal is free of infestation
- That thermal contrast identifies the cause
- That software can order an official quarantine
Where AI screening fits in the detection chain
No single method answers every screwworm question. The strongest operational design connects broad automated screening with close physical inspection and qualified confirmation.
| Method | Best use | Important limit |
|---|---|---|
| Fixed RGB + thermal chute screening | Repeatable, high-throughput review of every handled animal | Cannot see every hidden body area or confirm the parasite |
| Manual head-to-tail inspection | Close examination of navels, ears, folds, feet, and other hidden sites | Requires time, safe handling, trained attention, and repeated labor |
| Phone or close-up wound image | Documenting a concern after a person locates it | Capture timing, angle, lighting, and coverage vary by operator |
| Larval examination or molecular test | Species confirmation for an animal already considered suspicious | Starts after a wound or specimen has been found and collected |
Designed for feedlots, livestock markets, and cattle-handling facilities
For a commercial operation, the value is not another isolated camera feed. It is a review queue tied to normal cattle movement: which animal passed, whether complete views were captured, what changed, and which original frames need a person's attention.
Screen at handling speed
Collect consistent evidence during receiving, processing, loading, or scheduled health work.
Prioritize labor
Move the highest-risk passages to the front of the inspection queue instead of asking people to review every frame.
Preserve an audit trail
Keep time-stamped source imagery, capture quality, and escalation history attached to the animal record.
Built to be validated with researchers and animal-health teams
Researchers need more than a promising model score. A useful screwworm detection study needs expert-confirmed labels, negative controls, consistent sensor geometry, blinded evaluation, performance by wound location and coat condition, and prospective testing under real handling conditions.
Livestock Technologies is looking for university laboratories, veterinarians, diagnostic teams, livestock markets, and feedlots that can help design those protocols. The target is a defensible screening tool with published limitations—not a black-box disease label.
Validation required before a screwworm detection claim
The hard part is not drawing a box around a suspicious region. It is proving that the workflow performs reliably across coat colors, mud, rain, sun, motion, camera angles, wound types, and animals with no relevant lesion. Screwworm-specific validation also requires expert-confirmed ground truth; ordinary wound images are not a substitute.
Our release gate is therefore evidence, not a launch date: a documented dataset, blinded labels, capture-quality metrics, false-negative analysis, prospective site testing, and a clear escalation protocol with veterinary or animal-health partners. Until then, the website and product will say in development.
If a suspicious animal is found today
Follow the people-and-protocol path. USDA advises producers to watch for expanding or draining wounds, maggots or egg masses, foul odor, and pain or behavior change. Contact a veterinarian or animal-health official and use the current APHIS reporting instructions. Do not wait for a future AI feature, and do not use an image alone to confirm or dismiss a case.
Current detections and response information can change quickly. Check the USDA APHIS current-status page and reporting guidance for the latest instructions.
Frequently asked questions
How can AI help detect New World screwworm in cattle?
Computer vision can screen consistent RGB and thermal body views for visible wound-like changes, drainage, blood, fly activity, or localized heat patterns. It can prioritize an animal for immediate inspection, but a camera cannot confirm that screwworm larvae are present.
What is the Livestock Technologies Nose ID screwworm detection system?
It is an in-development cattle chute workflow that links the animal and loading record with controlled side-view RGB and thermal captures, image-quality checks, wound-risk flags, and a review packet for an operator, veterinarian, or animal-health official.
Can an AI cattle chute scanner confirm a screwworm infestation?
No. The system is designed for screening and escalation. Confirmation requires examination of the wound and identification of collected larvae by qualified animal-health personnel or an approved test.
Why use a fixed cattle chute instead of occasional phone photos?
A fixed chute provides a known path, repeatable distance, controlled lighting, multiple synchronized views, and a time-stamped record for every handled animal. Phone photos remain useful for close follow-up, but they depend on someone first finding and photographing the concern.
Is Nose ID screwworm screening available to feedlots today?
The multimodal screwworm-risk workflow is in development. Livestock Technologies is seeking feedlots, livestock markets, veterinary teams, and researchers for capture-design and prospective validation work.
What should a producer do after finding a suspicious wound or maggots?
Contact a veterinarian or animal-health official immediately and follow current USDA APHIS reporting instructions. Do not wait for software or use an image result to confirm or rule out screwworm.
Primary sources
- Screwworm.gov: current U.S. detections and response status
- USGS: rapid genetic testing and the role of species confirmation
- University of Arizona Cooperative Extension: livestock inspection guide
- USDA APHIS: investment in New World screwworm preparedness projects
- USDA APHIS: Grand Challenge recommended spending plan
This article is educational and describes research under development. It is not veterinary advice, a diagnostic tool, or a statement of USDA endorsement.