Source: PURDUE UNIVERSITY submitted to NRP
A NOVEL APPROACH TOWARD ADDRESSING KEEL BONE CONCERNS IN EXTENSIVE HOUSING
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1031573
Grant No.
2023-67021-41369
Cumulative Award Amt.
$591,500.00
Proposal No.
2022-11130
Multistate No.
(N/A)
Project Start Date
Sep 20, 2023
Project End Date
Sep 19, 2026
Grant Year
2023
Program Code
[A1521]- Agricultural Engineering
Recipient Organization
PURDUE UNIVERSITY
(N/A)
WEST LAFAYETTE,IN 47907
Performing Department
(N/A)
Non Technical Summary
Up to 73% of egg laying hens experience catastrophic and sometimes even fatal changes to their keel bone. Besides the humanitarian aspect of animal suffering, it has direct financial consequences for farmers and consumers alike.This research aims to develop and deploy new sensors, as well as construct an animal computer model, to (finally) determine the root causes of the problem. As expressed by our industry partners, the results of our investigations will be used to re-design bird housing, including changes to perch, nest and litter area(s), to minimize or eliminate the keel bone damage.
Animal Health Component
20%
Research Effort Categories
Basic
80%
Applied
20%
Developmental
0%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
31532102020100%
Goals / Objectives
Throughout the life of laying hens, 52-73% can undergo catastrophic and sometimes even fatal changes to their keel bone (KB). At present, there is insufficient knowledge of the root causes of KB developmental calamity. Currently there are limited ways to monitor the biomechanical conditions experienced by egg-laying hens. We seek to transform the assessment of KB damage by employing quantitative assessment methods enabled by leveraging sensor developments, computer simulation and tissue characterization, and to demonstrate the methodology to a critical example.The following are specific objectives of this project:Objective 1: Measure external loading on hens: We will employ additively manufactured, custom force sensor technology to enable direct measurements of external forces on birds from external loading (impact and body weight measured from skin mounted sensors).Objective 2: Measure internal loading and strain on keel bones: We will extend our sensor technology to directly measure deformation conditions from internal loading (egg laying). We will implement direct bone mounted force and strain sensors for attachment to the keel bone, to measure external forces (impact) and bone deformation (egg laying).Objective 3: The Virtual Bird Model: We will build the Virtual Bird Model (VBM) as a finite element model based on 3D imaging of bird anatomy and tissue biomechanical properties. We will employ VBM to compute the skeletal stress state under loading external and internal to the bird.Objective 4: Measure and evaluate keel bone damage in extensive housing considering stocking density: We will integrate sensing and the Virtual Bird Model. We then apply the integrated approach in a setting to birds housed in an enriched colony setting and use stocking density as the housing design variable. We thereby will extend the sensing approach to monitor multiple birds over an extended time period and simultaneously sense forces at external and internal locations on birds together with bone strains. We will apply VBM to analyze the interaction between birds and of birds with enrichments to quantify the potential for keel bone damage.
Project Methods
MEASURE EXTERNAL LOADING ON HENSThis study aims to collect force data from sensors on birds' skin to investigate keel bone fractures, particularly from high-force events like collisions. Custom-made PVdF force sensors, created via 3D printing, will be placed on the birds' skin over the keel bone. These sensors will be attached using medical adhesive tape or surgical glue. The collected force data will be processed by a compact development board equipped with customizable electronics for charge-to-voltage conversion, data storage, and wireless transmission. This board can be housed in a backpack between the bird's wings, and thin film wires will connect it to sensor sheet.To validate the data from previous experiment, external PVdF force sensors will be positioned around the keel bone area and subjected to known forces, ensuring the accuracy of force measurements. The collected force sensory data will undergo pre-processing and be stored locally on a specialized commercial board equipped with custom electronics, memory, and a wireless transmitter. This board will also incorporate additional sensors like an accelerometer and thermometer. All data will be timestamped, enabling insights such as correlations between measured forces and bird movements, including interactions with caretakers. Validation of timestamped data will involve room-mounted cameras capturing the bird's activities from various angles, similar to previous studies. The entire setup will be housed in a previously developed "backpack".MEASURE AND EVALUATE LOADING AND STRAIN ON KEEL BONESThe study's objective is to measure bone strains on the keel bone, complementing existing knowledge about applied forces. To achieve this, two strain sensors will be surgically attached to the keel bone, allowing for the measurement of strains resulting from external impacts (e.g., collisions) and internal forces (e.g., bird's weight or egg laying). Calibration of the strain sensor data will be conducted using 3D printed bone models and excised keel bones with known forces applied for validation.The study will use 3D printed keel bone models or keel bones from euthanized birds. Two strain sensors will be attached to the bone surface to measure strain in two directions. Controlled mechanical forces will be applied to the bone, and strain data will be collected from the sensors. Finite element models will analyze the bone, considering appropriate material properties. This analysis will provide theoretical strain data, and comparison between measured and computed strain data will serve as the basis for calibrating the sensor output.To measure the keel bone's actual strain under in-vivo conditions, surgical placement of strain sensors will be carried out. The collected data will be compared to calibration data to determine in-vivo strain levels. Similar to PVdF force sensors, a commercial board will be utilized to collect, store, and transmit this data, mounted on the birds' backs. Communication between the strain sensors and the board will be facilitated through thin film interconnects, minimizing bird discomfort and simplifying the surgery.Initially, thicker commercial strain sensors will be used for quicker data collection, but custom-designed, thinner sensors tailored to the keel bone's shape will be produced in-house. These custom sensors will offer greater flexibility and comfort for the birds, ultimately improving the quality of data collection and the birds' comfort during the experiment.The main goal is to accurately measure forces affecting birds' keel bones. Current 3D PVdF force sensors, while easy to use on the skin, do not directly measure bone forces, have uncertain lifespans, and their thickness can dampen impact measurements. To overcome these challenges, the study will develop nanomesh PVdF sensors that are surgically attached to the keel bone's surface, providing more precise and durable measurements. These nanomesh sensors, combined with ultra-thin strain sensors, are flexible and biocompatible, enabling long-term monitoring of both impact and egg-laying forces. The development of these sensors will be expedited by previous research efforts. Surgical placement of these sensors will be guided by an experienced poultry surgeon and overseen by a PACUC veterinarian, with data transmitted to a storage board located in a backpack on the bird's shoulder via thin film interconnects, minimizing bird discomfort during the experiment.VIRTUAL BIRD MODELThe study involves conducting 3D CT scans of 6 birds housed in conventional cages to analyze the biomechanics of their keel bones and pectoralis muscles. This imaging process follows established procedures and results in 3D representations of the birds' skeletal structures and muscle anatomy. Image segmentation techniques will be used to isolate the keel bone by analyzing image greyscales, allowing for the distinction between cartilage and bone domains, which are crucial for subsequent biomechanical analysis.The primary focus of the biomechanical characterization is on the keel bone, and this will be achieved using the in-vivo reference indentation method (RPI). Multiple RPI measurements will be conducted at various locations along the keel bone to understand its spatial variation in biomechanical properties. To validate this approach, reference samples from the keel bone will undergo both RPI and standard 3-point bend experiments. To characterize the pectoralis muscle, excised muscle tissue will be utilized due to limitations in current in-vivo methods. This will involve performing specialized biaxial tensile tests to measure hyperelastic passive tissue properties, with the resulting data being translated into elastic constitutive parameters following established protocols.The Virtual Bird Model (VBM) will be developed as a finite element model, utilizing reconstructed animal anatomy and specialized software to create a finite element mesh. It will incorporate tissue biomechanical properties and simulate deformations, strains, and stresses during quasi-static loading conditions resembling palpation exercises. Computational analysis will be performed using a finite element code, and the computed stress data for the keel bone will be compared to its strength data. A keel bone damage risk criterion will be calculated by assessing the ratio of computed principal stress / strain to measured bone strength. The VBM will undergo validation steps, including experiments on excised keel bones and extension to dynamic loading conditions, to assess its accuracy and effectiveness.EVALUATE KEEL BONE DAMAGE IN EXTENSIVE HOUSING CONSIDERING STOCKING DENSITYEgg-laying hens will be housed at varying stocking densities in enriched colony cage systems to demonstrate the proposed approach. Different cohorts of birds will be observed for varying durations, and data on body weight, egg production, and keel bone damage will be collected. All birds will wear acceleration sensors, and some focal hens will have custom-made PVdF sensors to monitor impact forces. The study aims to analyze acceleration data and video recordings to identify critical events, especially collisions with perches, and correlate them with stocking density. Researchers will also examine the relationship between impact forces, acceleration, and behavior, hypothesizing that significant accelerations correspond to high impact forces. Additionally, the study will calculate a new parameter, skeletal deformation distance, to describe a hen's ability to absorb energy during impacts, and measure bone strains to predict keel bone fracture likelihood. VBM will be used to analyze impact events, and its validity will be assessed by comparing computed contact forces and bone strain data to measured values in the enriched cage environment.

Progress 09/20/24 to 09/19/25

Outputs
Target Audience:The research is still in the preliminary stages of investigations. As such, we do not have any research findings that could be beneficial to a larger audience beyond the USDA/NIFA program managers. Changes/Problems:Personnel: the sensory data collection system is being developed by undergraduate students. Student(s) graduation necessitatestraining new students, including re-designing of some of the system components. To address some of these concerns, a graduate student was hired. However, as a result on some federal administration changes, that student was not able to join the project, which resulted in further developmental delays. On a technical side, a number of issues were identified, including changes to the piezoelectric force sensor signal conditioning circuit (from non-inverting voltage amplifier to differential charge amplifier) and strain gauge (increase resistance of wheatstone bridge to increase the signal resolution). What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?Nawrocki: We have presented the results of our development of piezoelectric force sensor and strain gauge at a materials science conference titled "European Materials Research Society", Fall 2025 meeting. What do you plan to do during the next reporting period to accomplish the goals?We are continuing to develop the force sensor and strain gauge to be able to measure forces impacted on the keel bone. We are also developing the sensory signal conditioning circuitry that will allow for the signal to be recorded. We are anticipating to be able to deploy the system on birds later this Fall 2025. The preliminary data, as collected from birds, will allow us to make further changes to the data collecting system.

Impacts
What was accomplished under these goals? Nawrocki: We are continuing to develop the portable poultry healthcare monitoring system. We have developed a system that can already include Inertial Measurement Unit (IMU, can collect movement of a bird along X, Y, and Z axis, along with rotation around X, Y, and Z)as well asa thermistor (to collect bird's temperature). We are still debugging the ability to collect force sensor (forces impacted at bird's keel bone) and strain sensor (deformation/expansion of the ribcage). We also have finished designing the ability to locally store the sensory data and offload it for future data processing. In a parallel track, we are continuing making progress to develop, calibrate, and characterize biocompatible piezoelectricforce sensors and strain gaugages that will be implanted underneath bird's skin. --------------- Siegmund:In the present reporting period a functional computational model for the analysis of skeletal stresses in egg-laying hens was established. This model transfers a CT scan image through image processing steps into a finite element model. Parts of the body beyond the torso are neglected. Soft tissue and hard tissue domains are segmented. Furthermore, rib joints are included. The keel bone receives specil attention and is segmented into an ossified and into a cartilage domain. Density characteristics of soft and hard tissue are assigned. The soft tissue is considered as a viscoelastic solid, the hard tissue is considered as an elastic solid with modulus data from experimental data. The body cavity is considered as a fluid filled cavity, and can include the egg as an additional model domain. The interaction of the body with the surrounding is accounted for by contact interactions. These contact interactions consider (1) the keel-external surface interaction and (2) the feather/skin-external surface interactions. The body is assigned an initial velocity boundary conditions. The model computes the interaction of the bird with either rigid planes or perches. The model is implemented in the commercial FE code ABAQUS and is solved with an explicit, time marching algorithm. Furthermore, a validation experiment was conducted with a pendulum impact tester and an instrumented force plate. This experiment measures the impact force-time history. The computational model of the bird is validated against the experimental data. In both cases, the force-time response deviates strongly from an idealized Hertz contact model. This finding justifies the development of a complex bird model. The impact force-time response history of the experiment exhibits an initial force peak, followed by a plateau domain terminated by the rebound of the bird. The computational model is capable to represent this finding with very good agreement. As the model is thus validated, it can now be used to compute the interaction of a bird with a perch in case of an impact resulting from a missed landing. In such a case the keel interacts strongly with the perch which leads to a computed acceleration magnitude in general agreement with observations (~10 x g, with max(a)=60 g). At the same time, stresses close to the impact site remain relatively low compared to what is expected as bone strength. On the other hand, significant tensile stresses are computed at the dorsal side of the keel. These tensile stresses are the result of the inertia imparted by the sudden stop of forward motion. The presence of an egg in the body cavity can exacerbate the dorsal side stress level.Considering that the cartilage-bone interface at ossification front is a weak spot in the skeleton, the dorsal side tensile stresses are interpreted to results in keel bone damage. --------------- Karcher: We continue to refine the use of Reference Point Indentation (RPI) to validate the use of determining skeletal parameters for informing the virtual bird. The surgical procedure continues to be developed and refined to ensure we are ready for that phase when it is time to execute the experiment.

Publications

  • Type: Other Status: Submitted Year Published: 2025 Citation: Fadayomi O. M., D.W. Seets, B.J. Emmert, S. Cloft, and D.M. Karcher. 2025. Reference point indentation correlates with bone mechanical properties in chicken tibia. Poult. Sci
  • Type: Other Status: Submitted Year Published: 2025 Citation: Emmert, B. J., D. W. Seets, S. Cloft, O. M. Fadayomi, and D. M. Karcher. 2025. Probing the validity of reference point indentation for detecting bone quality differences between caged and cage-free laying hens. Poult. Sci
  • Type: Conference Papers and Presentations Status: Other Year Published: 2025 Citation: Nawrocki, R.A., Fan, J., Newell, B., Bravo, J., Wu, W., 2025, Additively Manufactured Functional Materials with Embedded Sensing and Actuation, E-MRS Fall 2025


Progress 09/20/23 to 09/19/24

Outputs
Target Audience:The research is still in the preliminary stages of investigations. As such, we do not have any research findings that could be beneficial to a larger audience beyond the USDA/NIFA program managers. Changes/Problems:From previous section (what was accomplished): (1) As already mentioned, we have determined that the bird monitoring system needs to be redesigned to improve the accuracy of the strain recording. It appears that the raw data from Wheatstone bridge, with strain gauge, does not provide enough voltage change to accurately record the strain. We are planning on the use of Operational Amplifier(s) to provide the necessary voltage. (2) nothing to report at this time. (3) nothing to report at this time. What opportunities for training and professional development has the project provided?From previous section (what was accomplished): (1) The bird monitoring system is being designed and assembled by a group of undergraduate students, under the supervision of PD Nawrocki, as part of their Undergraduate Capstone Program. (2) Trained an undergraduate student this past year in an independent study and have had great discussion on the opportunities and challenges the data generated from the biodent will present to the team as we move forward. (3) Recruited, hired and trained a graduate student research assistant (MS level) to conduct finite element simulations of the stresses on the keel bone furing impact. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?From previous section (what was accomplished): (1) We have already determined that the bird monitoring system needs to be redesigned to improve the accuracy of the strain recording. Additional necessary changes will be found out during laboratory and field (on live birds) testing of the system. (2) We continued to refine the data generated from the biodent. (3) We seek to conduct further development of the virtual bird model. Computed data of accelerations will be compared to data from prior experiments conducted in the Karcher lab. Also,we seek to build an instrumented engineering equivalent of the virtual bird model and to construct an impact loading device. We thereby seek to measure impact-contact pressure transients. Such data can then be used to validate the computational model itself prior to modifying the model to consider actual hens.

Impacts
What was accomplished under these goals? There are three distinct aspects of the project: (1) development of a system that can monitor various parameters of the birds (namely body temperature, motion, forces impacted by the keel bone, and strain of the keel bone), (2) monitoring of the aforementioned bird parameters, and (3) the use of these parameters to evaluate the Virtual bird Model. The following are the updates of the three aspects. (1) We have finished a preliminary design of the system (v01 (large scale) and v02 (small scale, about the size of a smart watch, to be eventually fit onto a bird)), have assembled a prototype, and are currently testing its operation. The next iteration of the system (v03) will be redesigned to reflect "what was learned from v02". (2) We continued to refine the data generated from the biodent.Provided additional bird information to the engineering aspects as things continue to develop. (3) Developed Version 1 of the virtual bird model. The model uses STL files derived from CT scans of keel bones in conjunction with an abstracted torso geometry of the the hen. We have implemented the model into the finite element code ABAQUS and investigated several several approaches to the design of the finite element mesh. We conducted the first set of impact simulations against a rigid perch.

Publications

  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2024 Citation: Vyshniakova, K., Hosseini, M.J.M., Bai, H., Fatahi, M., Malacco, V.M.R., Donkin, S.S., Voyles, R.M., Zhang, H.H., Nawrocki, R.A., Aqueous Ammonia Sensor with Neuromorphic Detection, Advanced Electronic Materials, doi: 10.1002/aelm.202400509, 2024
  • Type: Other Status: Published Year Published: 2024 Citation: Emmert, B. J., T. H. Siegmund, G. S. Fraley, and D.M. Karcher. 2024. Cautions in interpretation of reference point indentation data. Poult. Sci. 103: (E-Suppl):121.
  • Type: Other Status: Published Year Published: 2024 Citation: Seets, D. W., S. Cloft, C. A. Kroger, B. J. Emmert, T. H. Siegmund and D. M. Karcher. 2024. Investigating the application of Reference Point Indentation as a possible tool to determine quality metrics for excised keel bones. Poult. Sci. 103: (E-Suppl):53.