Aristek Systems
Contact us
Preview of case

AI assistant for safer surgeries in vet clinics

A reputable U.S.-based veterinary clinic network offering comprehensive animal healthcare services across general surgery, preventive care, dental procedures, and emergency treatment.

Data science & AIVeterinary
Icon 1USA
Icon 2Veterinary
Icon 3Since 2022

  • Summary

    This case study explains how we built an AI assistant to support veterinarians during surgical procedures and patient management. The solution analyzes patient data, calculates anesthesia dosages, and provides real-time recommendations to help veterinary teams make safer and faster clinical decisions. In this material, we describe the challenges the clinic network faced, how the AI assistant was developed, and what results it delivered in daily veterinary practice.

    Key achievements

    Key achievements

    Image
    90%

    and more accuracy in anesthesia dosage calculations

    30%

    reduction in surgical prep time

    24%

    increase in vet team productivity

    Challenge

    The client, a veterinary clinic group with an existing SaaS platform, faced growing operational pressure as patient volume increased and procedures became more complex.

    • Inefficient anesthesia preparation

      Vets spent excessive time manually calculating dosages and cross-checking vitals, reducing time for high-value clinical work.

    • Fragmented patient data

      Clinicians navigated multiple apps to access lab results, medical history, and medications, slowing care and increasing the risk of errors.

    • Workflow bottlenecks

      When an emergency overlapped with routine surgeries, staff had no automated triage or reallocation. Imaging techs were idle while nurses juggled tasks, and one procedure was rushed, costing over an hour in coordination and raising the risk of error.

    • Care quality under growing demand

      Increasing patient volumes and more complex cases strained staff capacity, making it harder to ensure timely, precise care and minimize the risk of complications, while also putting additional pressure on staff.

      Despite the SaaS platform’s basic functionality, these challenges highlighted the need for AI-driven assistance to optimize workflows and improve treatment accuracy.

      Solution

      As part of our veterinary software solutions development services, we created an assistant for safer surgeries as a central component of a comprehensive SaaS platform.

      Integrated into a mobile app as well, our veterinary software solution uses real-time data and adaptive decision-making to help veterinarians make safer and faster decisions during surgeries and patient management – right at the point of care.

      • Icon of card 1

        Anesthesia dosage optimization

        The assistant, using machine learning (ML) technology, calculates anesthesia dosage using historical medical data, patient vitals, and medication profiles. It factors in unique patient traits (breed, age, previous surgeries, comorbidities) and adjusts dosages in real time as vitals change. This reduces dosage errors and adapts to patient responses mid-procedure.

      • Icon of card 2

        Pre-op risk assessment & triage

        Before surgery, the tool scans the patient’s history and current vitals to flag potential anesthesia risks like cardiac history or abnormal lab results and suggests additional tests if needed. It classifies surgical risk levels and provides confidence scores for proceeding.

      • Icon of card 3

        Real-time vitals interpretation

        During surgery, the ML-powered tool continuously monitors vital sign charts and identifies patterns that may indicate distress, such as potential hypoxia, arrhythmia, or anesthesia overdose. It alerts the vet team and recommends intervention steps.

      • Icon of card 4

        Personalized protocol generator

        Based on species, weight, prior medical history, and typical outcomes, the assistant creates personalized anesthetic protocols for each patient. These templates evolve over time as the system learns from historical success rates and updates best practices.

      • Icon of card 5

        Dental and wellness insights

        For dental procedures, the assistant interprets historical dental charts, predicts future risks (e.g., likelihood of periodontal disease), and offers preventive treatment recommendations based on breed and dietary history.

      • Icon of card 6

        Conversational AI interface

        An in-app chatbot allows the veterinary team to ask natural-language questions during procedures or case reviews, such as: ‘Show last 3 surgeries with similar vitals’ or ‘Show optimal anesthesia for an American Akita with low HGB.’ The assistant responds with clear, concise guidance backed by data.

      • Icon of card 7

        Offline predictive support

        Even without an internet connection, the platform provides surgical risk scores and dosage guidance using a condensed on-device model, ensuring uninterrupted use in field clinics or mobile operations.

        Project scope

        To build the smart vet assistant, available on both desktop and mobile, we took a phased approach that we always adhere to when building AI solutions for veterinary.

        Icon

        Discovery phase

        • Analyzed practitioner workflows, patient data flows, and operational bottlenecks through audits and interviews.
        • Mapped accuracy risks in anesthesia and inefficiencies in cross-team coordination into priority use cases.

        Data science

        • Defined AI applications across case management, anesthesia dosing, vitals monitoring, triage, and workflow optimization.
        • Built lightweight models for mobile-side calculations and larger predictive/monitoring models trained on AWS SageMaker.
        • Training data included vitals, species, medication protocols, and historical case outcomes.

        Frontend & UX

        • Embedded AI features into a mobile app for vets/nurses and two admin panels for managers and platform admins.
        • Designed smart templates, suggestion widgets, and monitoring alerts to blend seamlessly into clinical routines.

        Backend & cloud infrastructure

        • Deployed as a cloud-native service on AWS using ECS Fargate for elastic scaling, RDS for database reliability, and CloudFront for fast global access.
        • Blue-green deployments ensured safe, continuous updates.
        • Staged rollout to high-usage clinics with feedback loops for iterative improvements in accuracy and operational efficiency.

        How it works

        Veterinarians access the AI assistant through the mobile app, while clinic staff (nurses, technicians, administrators) use it via the admin panel of the SaaS platform. They can enter patient data, upload diagnostic results, or attach treatment histories directly within the existing clinic management system.

        Once the information is captured, the assistant analyzes vitals, lab results, past records, and prescribed medications to deliver tailored recommendations.

        That’s how the tool operates based on the data collected:

        • Evidence-backed safety

          Potential contraindications (e.g., abnormal hematocrit, drug interactions) are flagged automatically, ensuring no critical detail is overlooked.

        • Protocol generation

          Vet opens the in-app chat and types: “Generate anesthesia protocol for a 6-year-old American Akita, 35 kg, low HGB.” The assistant instantly produces a tailored protocol with safe dosages, highlighted risks, and monitoring guidance.

        • Post-op instructions

          The tool generates clear discharge notes in plain, client-friendly language, customized to procedure, medications, and language preferences.

        • Real-time support

          During surgery, the vet can ask follow-ups like: “What if HR drops below 70?” The assistant provides immediate, evidence-based adjustments.

        • Workflow integration

          The assistant compiles a complete patient snapshot (lab reports, vitals, imaging, and medication history). During emergencies, it also suggests triage order (e.g., stabilize trauma case first, postpone dental check) and flags overlapping schedules for faster coordination.

          Project screenshots

          • Image of slide 0
          • Image of slide 1
          • Image of slide 2

          Team

          • Image of slide 0

            Project Manager x1

          • Image of slide 1

            AI/ML Engineer x2

          • Image of slide 2

            Front-end developer x1

          • Image of slide 3

            Back-end developer x1

          • Image of slide 4

            QA Engineer x1

          • Image of slide 5

            UX/UI Designer x1

          Tools & Technologies

          Amazon SageMaker
          Elasticsearch
          GPT-4
          LlamaIndex
          PostgreSQL
          Python
          React
          TensorFlow
          TypeScript

          Project results

          • Over 12 months, the AI assistant cut surgical prep time by 30% and boosted veterinarian productivity by 24%, making daily operations smoother and less stressful.

          • Anesthesia dosing accuracy reached 97%, reducing the risk of complications in high-stakes surgeries with real-time calculations and alerts.

          • Vet clinics saw fewer post-op questions from clients, faster case turnaround, and a significant drop in anesthesia-related incidents, improving overall patient safety.

            The stakeholders' and users' feedback

            • With the AI looking over the data, we feel more in control. Surgeries run smoother, and our patients are safer because nothing slips through the cracks.”

              Veterinarians

            • Our teams are more efficient, and post-op data helps us improve protocols every month. The assistant has already saved several pets in tricky procedures.”

              Clinic manager

            • The instructions have become easier to understand, so we felt confident giving medication at home. Our cat recovered even faster than we expected!”

              Pet owners

            Insights

            • Icon of card 1

              Clinical adoption depends on trust in recommendations

              At first, some veterinarians treated the assistant as a second opinion rather than a primary decision tool. Over time, as the system consistently produced accurate dosage calculations and flagged potential risks, teams started relying on it more during surgical preparation.

            • Icon of card 2

              Workflow improvements mattered as much as AI accuracy

              The biggest efficiency gains did not come only from machine learning models. Bringing patient data, vitals, and lab results into a single interface reduced the time spent switching between systems and allowed clinicians to prepare for procedures faster.

            • Icon of card 3

              Early versions generated too many alerts

              During testing, the system sometimes flagged too many potential risks, which could distract surgeons during procedures. The models were adjusted to prioritize clinically relevant alerts and reduce unnecessary notifications.

              Key takeaways

              Key takeaways

              Image
              Icon 1

              Embed AI directly into existing workflows.
              Veterinary teams adopt AI solutions for veterinary faster when they are integrated into the systems they already use.

              Icon 2

              Prioritize clear and actionable recommendations.
              Medical teams benefit from concise guidance rather than complex technical outputs.

              Icon 3

              Design for reliability in different environments.
              Offline capabilities and mobile support are important for veterinary clinics and field operations.

              Icon 4

              Balance automation with human oversight.
              AI tools should support clinical decisions while keeping veterinarians in control of final treatment choices.

              From prep to post-op – make every step smarter with AI.

              Contact us to learn how we can improve your veterinary services with our smart veterinary software solutions.

              FAQ

              AI systems are designed to support veterinary teams by analyzing patient data, monitoring vitals, and suggesting possible actions. However, the final medical decisions always remain with the veterinarian.

              AI systems typically use patient medical records, vital signs, treatment protocols, medication data, and historical case outcomes. The quality and structure of this data strongly affect the accuracy of predictions.

              Yes. AI assistants can be integrated into existing veterinary management systems, mobile apps, or clinical software through APIs and shared databases, allowing clinics to improve workflows without replacing their current platforms.

              We use third-party cookies to improve your experience with aristeksystems.com and enhance our services. Click either 'Accept' or 'Manage' to proceed.