Multi-perspective region-based CNNs for vertebrae labeling in intraoperative long-length images

PURPOSE

Effective aggregation of intraoperative x-ray images that capture the patient anatomy from multiple view-angles has the potential to enable and improve automated image analysis that can be readily performed during surgery. We present multi-perspective region-based neural networks that leverage knowledge of the imaging geometry for automatic vertebrae labeling in Long-Film images - a novel tomographic imaging modality with an extended field-of-view for spine imaging.

METHOD

A multi-perspective network architecture was designed to exploit small view-angle disparities produced by a multi-slot collimator and consolidate information from overlapping image regions. A second network incorporates large view-angle disparities to jointly perform labeling on images from multiple views (viz., AP and lateral). A recurrent module incorporates contextual information and enforce anatomical order for the detected vertebrae. The three modules are combined to form the multi-view multi-slot (MVMS) network for labeling vertebrae using images from all available perspectives. The network was trained on images synthesized from 297 CT images and tested on 50 AP and 50 lateral Long-Film images acquired from 13 cadaveric specimens. Labeling performance of the multi-perspective networks was evaluated with respect to the number of vertebrae appearances and presence of surgical instrumentation.

RESULTS

The MVMS network achieved an F1 score of >96% and an average vertebral localization error of 3.3 mm, with 88.3% labeling accuracy on both AP and lateral images - (15.5% and 35.0% higher than conventional Faster R-CNN on AP and lateral views, respectively). Aggregation of multiple appearances of the same vertebra using the multi-slot network significantly improved the labeling accuracy (p < 0.05). Using the multi-view network, labeling accuracy on the more challenging lateral views was improved to the same level as that of the AP views. The approach demonstrated robustness to the presence of surgical instrumentation, commonly encountered in intraoperative images, and achieved comparable performance in images with and without instrumentation (88.9% vs. 91.2% labeling accuracy).

CONCLUSION

The MVMS network demonstrated effective multi-perspective aggregation, providing means for accurate, automated vertebrae labeling during spine surgery. The algorithms may be generalized to other imaging tasks and modalities that involve multiple views with view-angle disparities (e.g., bi-plane radiography). Predicted labels can help avoid adverse events during surgery (e.g., wrong-level surgery), establish correspondence with labels in preoperative modalities to facilitate image registration, and enable automated measurement of spinal alignment metrics for intraoperative assessment of spinal curvature.

Evaluating a dry vs. wet disinfection in boot baths on detection of porcine epidemic diarrhea virus and porcine reproductive and respiratory virus RNA

Maintaining biosecurity between swine barns is challenging, and boot baths are an easily implementable option some utilize to limit pathogen spread. However, there are concerns regarding their efficacy, especially when comparing wet or dry disinfectants. The objective of this study was to evaluate the efficacy of boot baths in reducing the quantity of detectable porcine epidemic diarrhea virus (PEDV) and porcine reproductive and respiratory syndrome virus (PRRSV) genetic material using wet or dry disinfectants. Treatments included 1) control, 2) dry chlorine powder (Traffic C.O.P., PSP, LLC, Rainsville, AL), and 3) wet quaternary ammonium/glutaraldehyde liquid (1:256 Synergize, Neogen, Lexington, KY). Prior to disinfection, rubber boots were inoculated with 1 mL of a co-inoculants of PRRSV (1×10 5 TCID50/mL) and PEDV (1×10 5 TCID50/mL) and dried for 15 minutes. After the drying period, a researcher placed the boot on the right foot and stepped directly on a stainless steel coupon (control). Alternatively, the researcher stepped first into a boot bath containing either the wet or dry sanitizer, stood for 3 s, and then stepped onto a steel coupon. After one minute, an environmental swab was then collected and processed from each boot and steel coupon. The procedure was replicated 12 times per disinfectant treatment. Samples were analyzed using a duplex qPCR at the Kansas State Veterinary Diagnostic Laboratory. Cycle threshold values were analyzed using SAS GLIMMIX v 9.4 (SAS, Inc., Cary, NC). There was no evidence of a disinfectant × surface × virus interaction (P &gt; 0.10). An interaction between disinfectant × surface impacted (P &lt; 0.05) the quantity of detectable viral RNA. As expected, the quantity of the viruses on the coupon were greatest in the control, indicating that a contaminated boot has the ability to transfer viruses from a contaminated surface to a clean surface. Comparatively, the dry disinfectant treatment resulted in no detectable viral RNA on either the boot or subsequent coupon. The wet disinfectant treatment had statistically similar (P &gt; 0.05) viral contamination to the control on the boot, but less viral contamination compared to the control on the metal coupon. In this experiment, a boot bath with dry powder was the most efficacious in reducing the detectable viral RNA on both boots and subsequent surfaces.

Joint synthesis and registration network for deformable MR-CBCT image registration for neurosurgical guidance

Objective.The accuracy of navigation in minimally invasive neurosurgery is often challenged by deep brain deformations (up to 10 mm due to egress of cerebrospinal fluid during neuroendoscopic approach). We propose a deep learning-based deformable registration method to address such deformations between preoperative MR and intraoperative CBCT.Approach.The registration method uses a joint image synthesis and registration network (denoted JSR) to simultaneously synthesize MR and CBCT images to the CT domain and perform CT domain registration using a multi-resolution pyramid. JSR was first trained using a simulated dataset (simulated CBCT and simulated deformations) and then refined on real clinical images via transfer learning. The performance of the multi-resolution JSR was compared to a single-resolution architecture as well as a series of alternative registration methods (symmetric normalization (SyN), VoxelMorph, and image synthesis-based registration methods).Main results.JSR achieved median Dice coefficient (DSC) of 0.69 in deep brain structures and median target registration error (TRE) of 1.94 mm in the simulation dataset, with improvement from single-resolution architecture (median DSC = 0.68 and median TRE = 2.14 mm). Additionally, JSR achieved superior registration compared to alternative methods-e.g. SyN (median DSC = 0.54, median TRE = 2.77 mm), VoxelMorph (median DSC = 0.52, median TRE = 2.66 mm) and provided registration runtime of less than 3 s. Similarly in the clinical dataset, JSR achieved median DSC = 0.72 and median TRE = 2.05 mm.Significance.The multi-resolution JSR network resolved deep brain deformations between MR and CBCT images with performance superior to other state-of-the-art methods. The accuracy and runtime support translation of the method to further clinical studies in high-precision neurosurgery.

Deformable MR-CT image registration using an unsupervised, dual-channel network for neurosurgical guidance

PURPOSE

The accuracy of minimally invasive, intracranial neurosurgery can be challenged by deformation of brain tissue - e.g., up to 10 mm due to egress of cerebrospinal fluid during neuroendoscopic approach. We report an unsupervised, deep learning-based registration framework to resolve such deformations between preoperative MR and intraoperative CT with fast runtime for neurosurgical guidance.

METHOD

The framework incorporates subnetworks for MR and CT image synthesis with a dual-channel registration subnetwork (with synthesis uncertainty providing spatially varying weights on the dual-channel loss) to estimate a diffeomorphic deformation field from both the MR and CT channels. An end-to-end training is proposed that jointly optimizes both the synthesis and registration subnetworks. The proposed framework was investigated using three datasets: (1) paired MR/CT with simulated deformations; (2) paired MR/CT with real deformations; and (3) a neurosurgery dataset with real deformation. Two state-of-the-art methods (Symmetric Normalization and VoxelMorph) were implemented as a basis of comparison, and variations in the proposed dual-channel network were investigated, including single-channel registration, fusion without uncertainty weighting, and conventional sequential training of the synthesis and registration subnetworks.

RESULTS

The proposed method achieved: (1) Dice coefficient = 0.82±0.07 and TRE = 1.2 ± 0.6 mm on paired MR/CT with simulated deformations; (2) Dice coefficient = 0.83 ± 0.07 and TRE = 1.4 ± 0.7 mm on paired MR/CT with real deformations; and (3) Dice = 0.79 ± 0.13 and TRE = 1.6 ± 1.0 mm on the neurosurgery dataset with real deformations. The dual-channel registration with uncertainty weighting demonstrated superior performance (e.g., TRE = 1.2 ± 0.6 mm) compared to single-channel registration (TRE = 1.6 ± 1.0 mm, p < 0.05 for CT channel and TRE = 1.3 ± 0.7 mm for MR channel) and dual-channel registration without uncertainty weighting (TRE = 1.4 ± 0.8 mm, p < 0.05). End-to-end training of the synthesis and registration subnetworks also improved performance compared to the conventional sequential training strategy (TRE = 1.3 ± 0.6 mm). Registration runtime with the proposed network was ∼3 s.

CONCLUSION

The deformable registration framework based on dual-channel MR/CT registration with spatially varying weights and end-to-end training achieved geometric accuracy and runtime that was superior to state-of-the-art baseline methods and various ablations of the proposed network. The accuracy and runtime of the method may be compatible with the requirements of high-precision neurosurgery.

Deformable 3D-2D registration for high-precision guidance and verification of neuroelectrode placement

Purpose.Accurate neuroelectrode placement is essential to effective monitoring or stimulation of neurosurgery targets. This work presents and evaluates a method that combines deep learning and model-based deformable 3D-2D registration to guide and verify neuroelectrode placement using intraoperative imaging.Methods.The registration method consists of three stages: (1) detection of neuroelectrodes in a pair of fluoroscopy images using a deep learning approach; (2) determination of correspondence and initial 3D localization among neuroelectrode detections in the two projection images; and (3) deformable 3D-2D registration of neuroelectrodes according to a physical device model. The method was evaluated in phantom, cadaver, and clinical studies in terms of (a) the accuracy of neuroelectrode registration and (b) the quality of metal artifact reduction (MAR) in cone-beam CT (CBCT) in which the deformably registered neuroelectrode models are taken as input to the MAR.Results.The combined deep learning and model-based deformable 3D-2D registration approach achieved 0.2 ± 0.1 mm accuracy in cadaver studies and 0.6 ± 0.3 mm accuracy in clinical studies. The detection network and 3D correspondence provided initialization of 3D-2D registration within 2 mm, which facilitated end-to-end registration runtime within 10 s. Metal artifacts, quantified as the standard deviation in voxel values in tissue adjacent to neuroelectrodes, were reduced by 72% in phantom studies and by 60% in first clinical studies.Conclusions.The method combines the speed and generalizability of deep learning (for initialization) with the precision and reliability of physical model-based registration to achieve accurate deformable 3D-2D registration and MAR in functional neurosurgery. Accurate 3D-2D guidance from fluoroscopy could overcome limitations associated with deformation in conventional navigation, and improved MAR could improve CBCT verification of neuroelectrode placement.

Sodium buffered formic acid concentration and feed pH is stable over a 3-month period

Promoting feed hygiene with organic acids is an effective method to prevent foodborne illnesses from bacterial infection. The stability and acidification of mash and pelleted feed with sodium buffered formic acid was investigated. The acid product was incorporated to reach total formate inclusion levels of 0, 6, or 12 g/kg for swine nursery feed; 0, 4, or 9 g/kg for swine finishing feed; and 0, 3, or 6 g/kg for broiler grower feed. Samples were analyzed for total formate and pH on d 4, 32, 60, or 88 post-manufacturing. The concentration of formate remained stable across an 88-d period (P < 0.01). Treatment with the formic acid product decreased feed pH with increasing inclusion levels (all P < 0.01). Within each inclusion level of acid and across time, pH tended to increase in pelleted feed and decrease in mash feeds (all P < 0.01); however, these changes were small (0.1 units pH). These data suggest that sodium buffered formic acid can be applied to both mash and pelleted feed to provide continuous acidification over a 3-month period.

3D vertebrae labeling in spine CT: an accurate, memory-efficient (Ortho2D) framework

Purpose.Accurate localization and labeling of vertebrae in computed tomography (CT) is an important step toward more quantitative, automated diagnostic analysis and surgical planning. In this paper, we present a framework (called Ortho2D) for vertebral labeling in CT in a manner that is accurate and memory-efficient.Methods. Ortho2D uses two independent faster R-convolutional neural network networks to detect and classify vertebrae in orthogonal (sagittal and coronal) CT slices. The 2D detections are clustered in 3D to localize vertebrae centroids in the volumetric CT and classify the region (cervical, thoracic, lumbar, or sacral) and vertebral level. A post-process sorting method incorporates the confidence in network output to refine classifications and reduce outliers. Ortho2D was evaluated on a publicly available dataset containing 302 normal and pathological spine CT images with and without surgical instrumentation. Labeling accuracy and memory requirements were assessed in comparison to other recently reported methods. The memory efficiency of Ortho2D permitted extension to high-resolution CT to investigate the potential for further boosts to labeling performance.Results. Ortho2D achieved overall vertebrae detection accuracy of 97.1%, region identification accuracy of 94.3%, and individual vertebral level identification accuracy of 91.0%. The framework achieved 95.8% and 83.6% level identification accuracy in images without and with surgical instrumentation, respectively. Ortho2D met or exceeded the performance of previously reported 2D and 3D labeling methods and reduced memory consumption by a factor of ∼50 (at 1 mm voxel size) compared to a 3D U-Net, allowing extension to higher resolution datasets than normally afforded. The accuracy of level identification increased from 80.1% (for standard/low resolution CT) to 95.1% (for high-resolution CT).Conclusions. The Ortho2D method achieved vertebrae labeling performance that is comparable to other recently reported methods with significant reduction in memory consumption, permitting further performance boosts via application to high-resolution CT.

Data-Driven Deformable 3D-2D Registration for Guiding Neuroelectrode Placement in Deep Brain Stimulation

PURPOSE

Deep brain stimulation is a neurosurgical procedure used in treatment of a growing spectrum of movement disorders. Inaccuracies in electrode placement, however, can result in poor symptom control or adverse effects and confound variability in clinical outcomes. A deformable 3D-2D registration method is presented for high-precision 3D guidance of neuroelectrodes.

METHODS

The approach employs a model-based, deformable algorithm for 3D-2D image registration. Variations in lead design are captured in a parametric 3D model based on a B-spline curve. The registration is solved through iterative optimization of 16 degrees-of-freedom that maximize image similarity between the 2 acquired radiographs and simulated forward projections of the neuroelectrode model. The approach was evaluated in phantom models with respect to pertinent imaging parameters, including view selection and imaging dose.

RESULTS

The results demonstrate an accuracy of (0.2 ± 0.2) mm in 3D localization of individual electrodes. The solution was observed to be robust to changes in pertinent imaging parameters, which demonstrate accurate localization with ≥20° view separation and at 1/10^th the dose of a standard fluoroscopy frame.

CONCLUSIONS

The presented approach provides the means for guiding neuroelectrode placement from 2 low-dose radiographic images in a manner that accommodates potential deformations at the target anatomical site. Future work will focus on improving runtime though learning-based initialization, application in reducing reconstruction metal artifacts for 3D verification of placement, and extensive evaluation in clinical data from an IRB study underway.

The development of a noninvasive behavioral test for assessment of goat-human interactions

The objective was to develop a protocol for a noninvasive behavioral test to assess goat-human interactions. Boer goats (n = 45) were housed in groups of 3 at the K-State Sheep and Meat Goat Unit. A 3-min human approach test (HAT) was conducted after their first- and seventh-handling experience (i.e., moved by group through the chute and scale). Video footage was analyzed using specialized software (Observer 11.5 XT, Leesburg, VA, USA). The ethogram included three main categories (mutually exclusive within category): 1) spatial (close, middle, far); 2) orientation (facing vs. turned-away); and 3) structural (lie, stand, and nutritive and nonnutritive oral behaviors). The durations of these behavior outcomes were converted to percent, and then applied to a weighted formula to create an approach index (AI). This index placed behaviors on a 0 to 100% scale (0 = lying in back of the pen; 100 = closest to the human, performing oral behaviors). The Proc Univariate (SAS v.9.3, SAS Inst. Inc., Cary, NC, USA) was used to evaluate descriptive statistics and calculate sample size for future projects from all behavior outcomes. In addition, each goat AI categorized as great approach (GA; ≥75% quartile), moderate approach (MA; 25% to 75% quartiles), or least approach (LA; ≤ 25%) and the 1st vs. 7th handling experience was evaluated using Chi-square (χ ²) analysis. The CV% was low for the AI outcome; therefore, less animals are needed if the AI is used rather than other behavior outcomes used. After the first-handling experience, there were less GA-goats than what was expected from the χ ² analysis (χ ² = 17.6; P ≤ 0.01; residual = 0.26; expected = 24.5). After the seventh-handling experience, there were more MA-goats than expected (χ ² = 17.6; P ≤ 0.01; residual = 1.92; expected = 52.0). Repeated handling appeared to moderately improve goat approach to humans, which indicates that this protocol is a promising behavioral test to assess welfare of goats.

Data-Driven Detection and Registration of Spine Surgery Instrumentation in Intraoperative Images

PURPOSE

Conventional model-based 3D-2D registration algorithms can be challenged by limited capture range, model validity, and stringent intraoperative runtime requirements. In this work, a deep convolutional neural network was used to provide robust initialization of a registration algorithm (known-component registration, KC-Reg) for 3D localization of spine surgery implants, combining the speed and global support of data-driven approaches with the previously demonstrated accuracy of model-based registration.

METHODS

The approach uses a Faster R-CNN architecture to detect and localize a broad variety and orientation of spinal pedicle screws in clinical images. Training data were generated using projections from 17 clinical cone-beam CT scans and a library of screw models to simulate implants. Network output was processed to provide screw count and 2D poses. The network was tested on two test datasets of 2,000 images, each depicting real anatomy and realistic spine surgery instrumentation - one dataset involving the same patient data as in the training set (but with different screws, poses, image noise, and affine transformations) and one dataset with five patients unseen in the test data. Assessment of device detection was quantified in terms of accuracy and specificity, and localization accuracy was evaluated in terms of intersection-over-union (IOU) and distance between true and predicted bounding box coordinates.

RESULTS

The overall accuracy of pedicle screw detection was ~86.6% (85.3% for the same-patient dataset and 87.8% for the many-patient dataset), suggesting that the screw detection network performed reasonably well irrespective of disparate, complex anatomical backgrounds. The precision of screw detection was ~92.6% (95.0% and 90.2% for the respective same-patient and many-patient datasets). The accuracy of screw localization was within 1.5 mm (median difference of bounding box coordinates), and median IOU exceeded 0.85. For purposes of initializing a 3D-2D registration algorithm, the accuracy was observed to be well within the typical capture range of KC-Reg.¹.

CONCLUSIONS

Initial evaluation of network performance indicates sufficient accuracy to integrate with algorithms for implant registration, guidance, and verification in spine surgery. Such capability is of potential use in surgical navigation, robotic assistance, and data-intensive analysis of implant placement in large retrospective datasets. Future work includes correspondence of multiple views, 3D localization, screw classification, and expansion of the training dataset to a broader variety of anatomical sites, number of screws, and types of implants.

Consumer Sensory Evaluation of Beef Top Sirloin Cap Steaks from Four USDA Quality Grades

ObjectivesThe objective of this study was to evaluate the influence of quality grade on the palatability of beef top sirloin cap (biceps femoris) steaks.Materials and MethodsFour quality treatments [Prime, Top Choice (Modest and Moderate marbling), Low Choice and Select] were equally represented (n = 15/treatment) from beef top sirloin caps (IMPS # 184D). Sirloin caps were fabricated into 2.5 cm steaks from posterior to anterior following a 28-d aging period and randomly assigned to one of 3 analysis methods: Warner-Bratzler shear force (WBSF), fat and moisture analysis, and consumer sensory analysis. Steaks were cooked on a clamshell grill (Cuisinart Gridler Deluxe, Model GR-150, East Windsor, NJ) to a peak medium (71°C) degree of doneness monitored using a thermometer (Super-Fast Thermopen, ThermoWorks, American Fork, UT). Consumers (N = 118) evaluated each sample for juiciness, tenderness, flavor liking, and overall liking on a 0 to 100-point continuous line scales. Additionally, consumers rated each trait as either acceptable or unacceptable and classified all samples as one of 4 quality levels: unsatisfactory, everyday quality, better than everyday quality, or premium quality. Data were analyzed as a completely randomized design with the fixed effect of quality treatment.ResultsConsumers rated Top Choice, Low Choice, and Select similar (P < 0.05) for overall like, however, Prime rated (P < 0.05) higher than all other treatments. Also, Prime and Top Choice were similar (P > 0.05) for flavor liking, with Low Choice and Select also similar to Top Choice (P > 0.05). There was no difference (P > 0.05) among the quality treatments for tenderness and juiciness ratings. Similar to the rating results, when evaluating the percentage of samples rated acceptable for each palatability trait, no differences (P > 0.05) were found among quality treatments for tenderness, juiciness, and flavor, with all traits rated over 71.5% acceptable. However, a greater (P < 0.05) percentage of Prime samples were rated acceptable overall compared to Low Choice and Select. Additionally, there was no difference (P > 0.05) among the quality treatments for the percentage of samples classified as unsatisfactory. Consumers perceived a similar (P > 0.05) percentage of Top Choice and Low Choice samples at each quality level. Moreover, Prime had a greater percentage (P < 0.05) of samples perceived as Premium Quality than Select. For WBSF, there were no differences (P > 0.05) among treatments. Prime steaks had a similar (P > 0.05) moisture percentage as all other treatments, with Select having the greatest (P < 0.05) percentage of moisture compared to Top Choice and Low Choice. Furthermore, Top Choice and Low Choice had a similar (P > 0.05) percentage of fat, with Prime having the highest (P < 0.05) fat percentage and Select having the lowest (Prime > Top Choice = Low Choice > Select).ConclusionThese results indicate that quality grade has minimal impact on the palatability of beef top sirloin cap steaks. Therefore, food service does not need to pay the extra premiums associated with a higher grading product, as consumers will experience the same eating experience as with lower quality grades.

The amino acid composition and protein quality of various egg, poultry meal by-products, and vegetable proteins used in the production of dog and cat diets

New protein ingredients are used to support pet food market growth and the development of new products while maintaining animal dietary needs. However, novel protein sources (e.g., spray-dried chicken, and (or) rice, pea, and potato protein concentrates) have limited data available regarding their protein quality. The objective of this study was to evaluate protein ingredients used in the pet food industry by laboratory analysis and a chick growth assay as a model. Following analysis for proximate and amino acid composition, chicks (six birds per pen with four pens per treatment) were fed experimental diets for 10 d. Diets contained 10% crude protein from each of the experimental protein sources (spray-dried egg-SDEG; spray-dried egg white-SDEW, spray-dried inedible whole egg-SDIE, chicken by-product meal-CBPM, chicken meal-CKML, low-temperature fluid bed air-dried chicken-LTCK, low-temperature and pressure fluid bed dried chicken-LTPC, spray-dried chicken-SDCK, whey protein concentrate-WPCT, corn gluten meal-CGML, corn protein concentrate-CPCT, potato protein isolate-PPIS, rice protein concentrate-RPCT, pea protein isolate-PEPI, soy protein isolate-SPIS, and soybean meal-SBML) along with an N-free diet (negative control). Chicks fed SDEG, SDIE, and LTPC had the highest protein efficiency ratio (PER; P < 0.0001; 5.18, 5.37, and 5.33, respectively), LTCK and SDCK were intermediate (4.54 and 4.79), and the CBPM and CKML were the lowest among the poultry proteins for EAA:NEAA, PER, and Lys availability. Among the vegetable proteins, PPIS and SBML had the highest PER values (3.60 and 3.48, P < 0.0001). In general, the chick PER method ranked the quality of animal protein sources higher than vegetable proteins, and these results were consistent with the EAA:NEAA ratio and Lys availability.

Impact of including distillers' dried grains with solubles at expense of soybean meal on Boer-influenced goat growth performance

The objective of this project was to evaluate the efficacy of distillers' dried grains (DDGS) as a replacement for soybean meal (SBM) in a diet for Boer-influenced goats. Because diets were isonitrogenous and isocaloric, it was hypothesized that the replacement of SBM with DDGS would reduce diet costs without affecting growth performance. To test this hypothesis, forty-eight Boer-influenced goats (28.2 ± 0.96 kg BW; approximately 90 d of age) were allocated to four treatments in a completely random design in a climate-controlled (13°C) facility of the Kansas State University Sheep and Meat Goat Center. Goats were stratified by BW into 16 pens (four pens per treatment; three goats per pen) for a 42-d finishing study. The main effect concentration of DDGS (0%, 10%, 20%, or 30% DM basis) which resulted in dietary treatments of the following: 1) 0% SBM replaced by DDGS (0DDGS); 2) 33% SBM replaced by DDGS (10DDGS); 3) 66% SBM replaced by DDGS (20DDGS); and 4) 100% SBM replaced by DDGS (30DDGS). Goats remained on a self-feeder with continuous access to their respective pelleted, complete dietary treatments, and clean, fresh water. Two-day weights were taken at the beginning and end of the trial, with BW, ADG, ADFI, and G:F measured and calculated every 7 d. Beginning BW were similar across all treatments (P = 0.99). The inclusion of DDGS had no effect on BW or ADFI from days 0 to 42 (P ≥ 0.15). Both ADG (P = 0.04) and G:F (P = 0.001) increased linearly with increasing DDGS inclusion, with goats fed diets containing DDGS having 33% greater (P = 0.05) ADG than those fed diets without DDGS. Ultimately, these results confirm the hypothesis that DDGS can be used to replace SBM in the diet. At current commodity prices, this leads to a 17% savings in diet cost. This novel research shows similar improved performance measures to lambs and cattle when feeding DDGS.

Effects of dietary supplementation of formaldehyde and crystalline amino acids on gut microbial composition of nursery pigs

Formaldehyde-based feed additives are approved in the US for Salmonella control and reducing bacterial contamination in animal feed. However, we hypothesize formaldehyde inclusion in swine diets may influence gut microbial composition due to its antimicrobial properties which might negatively influence microbial populations and pig growth performance. Also, formaldehyde inclusion in diets is known to reduce the dietary availability of amino acids. Therefore, our study was conducted to characterize if the effects of feed formaldehyde-treatment are due to influences on microbial population or diet amino acid (AA) sources. Dietary treatments were arranged in a (2 × 2) + 1 factorial with formaldehyde treatment (none vs. 1000 ppm formaldehyde) and crystalline AA inclusion (low vs. high) with deficient AA content plus a positive control diet to contain adequate AA content without dietary formaldehyde. Treating diets with formaldehyde reduced growth rate (P = 0.001) while the AA inclusion had no evidence of impact. Formaldehyde reduced feed bacterial content and altered fecal microbial communities (P < 0.05). Therefore, we conclude that the negative influence on growth was due to the impact on the fecal microbial community. Implications are that strategies for feed pathogen control need to take into account potential negative impacts on the gut microbial community.

Stability of four commercial phytase products under increasing thermal conditioning temperatures

Phytase is a feed-grade enzyme frequently added to swine diets to help improve the digestibility of phytate phosphorus. However, like any enzyme, it may be subject to heat damage when exposed to thermal processing. Therefore the objective of this experiment was to determine the stability of 4 commercial phytase products exposed to increasing thermal conditioning temperatures in the pelleting process. The 4 commercial products used were: Quantum Blue G (AB Vista, Plantation, FL); Ronozyme Hi Phos GT (DSM Nutritional Products, Parsippany, NJ); Axtra Phy TPT (Dupont, Wilmington, DE), and Microtech 5000 Plus (Guangdong Vtr Bio-Tech Co., Ltd., Guangdong, China). The phytase products were mixed as part of a corn-soybean meal-based swine diet at a concentration recommended by the manufacturer to provide a 0.12% aP release. Diets were exposed to each of 4 thermal conditioning temperatures (65, 75, 85, and 95°C) and the entire process repeated on 4 consecutive days to create 4 replicates. Samples were taken while feed exited the conditioner and before entering the pellet die. Samples were cooled to room temperature before being stored in plastic bags until analysis. Phytase stability was measured as the residual phytase activity (% of initial) at each conditioning temperature. There were no product × temperature interactions observed for conditioning temperature, conditioner throughput, or residual phytase activity. As target temperature increased, conditioner throughput decreased (linear; P < 0.001) and phytase activity decreased (linear; P < 0.001) for each product. Residual phytase activity decreased as conditioning temperature increased from 65 to 95°C at a rate of -1.9% for every 1°C increase in conditioning temperature. There was a significant phytase product (P < 0.001) main effect which was mainly driven by Microtech 5000 Plus having decreased (P < 0.05) phytase activity when compared to all other products at 65, 75, and 85°C. However at 95°C Axtra Phy TPT had greater (P < 0.05) residual phytase activity compared with Microtech 5000 Plus, with Quantum Blue G and Ronozyme Hi Phos intermediate. Increasing target conditioning temperatures decreased phytase stability regardless of product. In addition, Microtech 5000 Plus had decreased residual phytase activity (% of initial) when compared to all other products at 65, 75, and 85°C.

Effect of pelleting on survival of porcine epidemic diarrhea virus-contaminated feed

Porcine epidemic diarrhea virus (PEDV) is a heat-sensitive virus that has devastated the U.S. swine industry. Because of its heat sensitivity, we hypothesized that a steam conditioner and pellet mill mimicking traditional commercial thermal processing may mitigate PEDV infectivity. Pelleting, a common feed processing method, includes the use of steam and shear forces, resulting in increased temperature of the processed feed. Two thermal processing experiments were designed to determine if different pellet mill conditioner retention times and temperatures would impact PEDV quantity and infectivity by analysis of quantitative reverse transcription PCR and bioassay. In Exp. 1, a 3 × 3 × 2 factorial design was used with 3 pelleting temperatures (68.3, 79.4, and 90.6°C), 3 conditioning times (45, 90, or 180 s), and 2 doses of viral inoculation (low, 1 × 10 tissue culture infectious dose (the concentration used to see cytopathic effect in 50% of the cells)/g, or high, 1 × 10 tissue culture infectious dose/g). Noninoculated and PEDV-inoculated unprocessed mash were used as controls. The low-dose PEDV-infected mash had 6.8 ± 1.8 cycle threshold (Ct) greater ( < 0.05) PEDV than the high-dose mash. Regardless of time or temperature, pelleting reduced ( < 0.05) the quantity of detectable viral PEDV RNA compared with the PEDV-inoculated unprocessed mash. Fecal swabs from pigs inoculated with the PEDV-positive unprocessed mash, regardless of dose, were clinically PEDV positive from 2 to 7 d (end of the trial) after inoculation. However, if either PEDV dose of inoculated feed was pelleted at any of the 9 tested conditioning time × temperature combinations, no PEDV RNA was detected in fecal swabs or cecum content. Based on Exp. 1 results, a second experiment was developed to determine the impact of lower processing temperatures on PEDV quantity and infectivity. In Exp. 2, PEDV-inoculated feed was pelleted at 1 of 5 conditioning temperatures (37.8, 46.1, 54.4, 62.8, and 71.1°C) for 30 s. The 5 increasing processing temperatures led to feed with respective mean Ct values of 32.5, 34.6, 37.0, 36.5, and 36.7, respectively. All samples had detectable PEDV RNA. However, infectivity was detected by bioassay only in pigs from the 37.8 and 46.1°C conditioning temperatures. Experiment 2 results suggest conditioning and pelleting temperatures above 54.4°C could be effective in reducing the quantity and infectivity of PEDV in swine feed. However, additional research is needed to prevent subsequent recontamination after pelleting as it is a point-in-time mitigation step.

Impact of varying analytical methodologies on grain particle size determination

The determination of particle size is an important quality control measurement for feed manufacturers, nutritionists, and producers. The current approved method for determining the geometric mean diameter by weight (d) and geometric standard deviation (S) of grains is standard ANSI/ASAE S319.4. This method controls many variables, including the suggested quantity of initial material and the type, number, and size of sieves. However, the method allows for variations in sieving time, sieve agitators, and the use of a dispersion agent. The objective of this experiment was to determine which method of particle size analysis best estimated the particle size of various cereal grain types. Eighteen samples of either corn, sorghum, or wheat were ground and analyzed using different variations of the approved method. Treatments were arranged in a 5 × 3 factorial arrangement with 5 sieving methods: 1) 10-min sieving time with sieve agitators and no dispersion agent, 2) 10-min sieving time with sieve agitators and dispersion agent, 3) 15-min sieving time with no sieve agitators or dispersion agent, 4) 15-min sieving time with sieve agitators and no dispersion agent, and 5) 15-min sieving time with sieve agitators and dispersion agent conducted in 3 grain types (ground corn, sorghum, and wheat) with 4 replicates per treatment. The analytical method that resulted in the lowest d and greatest S was considered desirable because it was presumably representative of increased movement of particles to their appropriate sieve. Analytical method affected d and S ( ≤ 0.05) measured by both standards. Inclusion of sieve agitators and dispersion agent in the sieve stack resulted in the lowest d, regardless of sieving time. Inclusion of dispersion agent reduced d ( ≤ 0.05) by 32 and 36 µm when shaken for 10 and 15 min, respectively, compared to the same sample analyzed without dispersion agent. The addition of the dispersion agent also increased S. The dispersion agent increased the quantity of very fine particles collected in the pan; therefore, S was significantly greater ( ≤ 0.05). Corn and sorghum ground using the same mill parameters had similar d ( > 0.05), but wheat ground using the same mill parameters was 120 to 104 µm larger ( ≤ 0.05) than corn and sorghum, respectively. Both sieve agitators and dispersion agent should be included when conducting particle size analysis. The results indicate that 10 and 15 min of sieving time produced similar results.

Effects of wheat source and particle size in meal and pelleted diets on finishing pig growth performance, carcass characteristics, and nutrient digestibility

Two experiments were conducted to test the effects of wheat source and particle size in meal and pelleted diets on finishing pig performance, carcass characteristics, and diet digestibility. In Exp. 1, pigs (PIC 327 × 1050; = 288; initially 43.8 kg BW) were balanced by initial BW and randomly allotted to 1 of 3 treatments with 8 pigs per pen (4 barrows and 4 gilts) and 12 pens per treatment. The 3 dietary treatments were hard red winter wheat ground with a hammer mill to 728, 579, or 326 μm, respectively. From d 0 to 40, decreasing wheat particle size decreased (linear, < 0.033) ADFI but improved (quadratic, < 0.014) G:F. From d 40 to 83, decreasing wheat particle size increased (quadratic, < 0.018) ADG and improved (linear, < 0.002) G:F. Overall from d 0 to 83, reducing wheat particle size improved (linear, < 0.002) G:F. In Exp. 2, pigs (PIC 327 × 1050; = 576; initially 43.4 ± 0.02 kg BW) were used to determine the effects of wheat source and particle size of pelleted diets on finishing pig growth performance and carcass characteristics. Pigs were randomly allotted to pens, and pens of pigs were balanced by initial BW and randomly allotted to 1 of 6 dietary treatments with 12 replications per treatment and 8 pigs/pen. The experimental diets used the same wheat-soybean meal formulation, with the 6 treatments using hard red winter or soft white winter wheat that were processed to 245, 465, and 693 μm and 258, 402, and 710 μm, respectively. All diets were pelleted. Overall, feeding hard red winter wheat increased ( < 0.05) ADG and ADFI when compared with soft white winter wheat. There was a tendency ( < 0.10) for a quadratic particle size × wheat source interaction for ADG, ADFI, and both DM and GE digestibility, as they were decreased for pigs fed 465-μm hard red winter wheat and were greatest for pigs fed 402-μm soft white winter wheat. There were no main or interactive effects of particle size or wheat source on carcass characteristics. In summary, fine grinding hard red winter wheat fed in meal form improved G:F and nutrient digestibility, whereas reducing particle size of wheat from approximately 700 to 250 μm in pelleted diets did not influence growth or carcass traits. Finally, feeding hard red winter wheat improved ADG and ADFI compared with feeding soft white winter wheat.

Quantitative Susceptibility Mapping Suggests Altered Brain Iron in Premanifest Huntington Disease

BACKGROUND AND PURPOSE

In patients with premanifest (nonsymptomatic) and advanced Huntington disease, changes in brain iron levels in the basal ganglia have been previously reported, especially in the striatum. Quantitative susceptibility mapping by using MR phase imaging allows in vivo measurements of tissue magnetic susceptibility, which has been shown to correlate well with iron levels in brain gray matter and is believed to be more specific than other imaging-based iron measures. The purpose of this study was to investigate the use of magnetic susceptibility as a biomarker of disease progression.

MATERIALS AND METHODS

Fifteen subjects with premanifest Huntington disease and 16 age-matched healthy controls were scanned at 7T. Magnetic susceptibility, effective relaxation, and tissue volume in deep gray matter structures were quantified and compared with genetic and clinical measures.

RESULTS

Subjects with premanifest Huntington disease showed significantly higher susceptibility values in the caudate nucleus, putamen, and globus pallidus, indicating increased iron levels in these structures. Significant decreases in magnetic susceptibility were found in the substantia nigra and hippocampus. In addition, significant volume loss (atrophy) and an increase effective relaxation were observed in the caudate nucleus and putamen. Susceptibility values in the caudate nucleus and putamen were found to be inversely correlated with structure volumes and directly correlated with the genetic burdens, represented by cytosine-adenine-guanine repeat age-product-scaled scores.

CONCLUSIONS

The significant magnetic susceptibility differences between subjects with premanifest Huntington disease and controls and their correlation with genetic burden scores indicate the potential use of magnetic susceptibility as a biomarker of disease progression in premanifest Huntington disease.