An official website of the United States government
Leptospirosis is a life-threatening, zoonotic disease with various clinical presentations, including renal injury, hepatic injury, pancreatitis, and pulmonary hemorrhage. With prompt recognition of the disease and treatment, 90% of infected dogs have a positive outcome. Therefore, rapid, early diagnosis of leptospirosis is crucial. Testing for Leptospira-specific serum antibodies using the microscopic agglutination test (MAT) lacks sensitivity early in the disease process, and diagnosis can take >2 wk because of the need to demonstrate a rise in titer. We applied machine-learning algorithms to clinical variables from the first day of hospitalization to create machine-learning prediction models (MLMs). The models incorporated patient signalment, clinicopathologic data (CBC, serum chemistry profile, and urinalysis = blood work [BW] model), with or without a MAT titer obtained at patient intake (=BW + MAT model). The models were trained with data from 91 dogs with confirmed leptospirosis and 322 dogs without leptospirosis. Once trained, the models were tested with a cohort of dogs not included in the model training (9 leptospirosis-positive and 44 leptospirosis-negative dogs), and performance was assessed. Both models predicted leptospirosis in the test set with 100% sensitivity (95% CI: 70.1–100%). Specificity was 90.9% (95% CI: 78.8–96.4%) and 93.2% (95% CI: 81.8–97.7%) for the BW and BW + MAT models, respectively. Our MLMs outperformed traditional acute serologic screening and can provide accurate early screening for the probable diagnosis of leptospirosis in dogs.
more »
« less
This content will become publicly available on June 22, 2026
Performance of a machine learning model for real-time prediction of health status of lactating cows
The objective of this study was to evaluate the performance of an XGBoost model trained with behavioral, physiological, performance, environmental, and cow feature data for classifying cow health status (HS). The model predicted HS based on physical activity, resting, reticulo-rumen temperature, rumination and eating behavior, milk yield, conductivity and components, temperature and humidity index, parity, calving features, and stocking density. Daily at 5 a.m., the model generated a HS prediction [0 = no health disorder (HD); 1 = health disorder]. At 7 a.m., technicians blind to the prediction conducted clinical exams on cows from 3 to 11 DIM to classify cows (n = 625) as affected (HD = 1) or not (HD = 0) by metritis, mastitis, ketosis, indigestion, displaced abomasum, and pneumonia. Using each day a cow presented clinical signs of HD as a positive case (i.e., HD = 1), metrics of performance (%; 95% CI) were: sensitivity (Se) = 57 [52, 62], specificity = 81 [80, 82]; positive predictive value (PPV) = 20 [18, 22], negative predictive value = 96 [95, 96], accuracy = 79 [78, 80], balanced accuracy = 69 [66, 72], F-1 Score = 29 [26, 32]. Sensitivity was also evaluated using fixed time intervals around clinical diagnosis of disease as a positive case (Table 1). Our findings suggest that the ability of an XGBoost algorithm trained on diverse sensor and nonsensor data to identify cows with HD was moderate when only days when cows presented clinical signs of disease were considered a positive case. Sensitivity and PPV can be improved substantially when all days within fixed intervals before and after clinical diagnosis are used as positive cases. Table 1 (Abstr. 2614). Sensitivity and PPV for an XGBoost algorithm trained to predict cow health status using fixed intervals before and after clinical diagnosis as positive cases Day relative to CD Se (%) 95% CI PPV (%) 95% CI −5 to 0 58 49, 67 21 16, 25 −3 to 0 55 46, 64 19 15, 24 −5 to 1 69 61, 78 24 20, 29 −5 to 3 81 73, 88 28 23, 33 −5 to 5 86 80, 92 30 25, 34 −3 to 1 67 58, 75 23 18, 27 −3 to 3 78 70, 86 27 22, 31 −3 to 5 83 76, 90 28 24, 33 0 to 3 75 68, 83 24 20, 29 0 to 5 81 73, 88 26 21, 31 −1 to 0 54 44, 63 18 14, 22 0 to 1 63 54, 72 20 16, 25 −1 to 1 66 57, 75 21 17, 26
more »
« less
- Award ID(s):
- 2211941
- PAR ID:
- 10633087
- Publisher / Repository:
- American Dairy Science Association J. Dairy Sci. 108 (E-Suppl. 1):370. (Abstr)
- Date Published:
- Volume:
- 108
- Issue:
- E-Supplement 1
- Page Range / eLocation ID:
- 370
- Format(s):
- Medium: X
- Location:
- chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.adsa.org/Portals/0/SiteContent/Docs/Meetings/2025ADSA/Abstracts_BOOK_2025_20250624-1249.pdf
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Chen, Nan-Hua (Ed.)Background:Although the rate of emerging infectious diseases that originate in wildlife has been increasing globally in recent decades, there is currently a lack of epidemiological data from wild animals. Methodology:We used serology to determine prior exposure to foot‐and‐mouth disease virus (FMDV),Brucellaspp., andCoxiella burnetiiand used genetic testing to detect blood‐borne parasitic infections in the generaEhrlichia,Anaplasma,Theileria, andBabesiafrom wildlife in two national parks, Kruger National Park (KNP), South Africa, and Etosha National Park (ENP), Namibia. Serum and whole blood samples were obtained from free‐roaming plains zebra (Equus quagga), greater kudu (Tragelaphus strepsiceros), impala (Aepyceros melampus), and blue wildebeest (Connochaetes taurinus). Risk factors (host species, sex, and sampling park) for infection with each pathogen were assessed, as well as the prevalence and distribution of co‐occurring infections. Results:In KNP 13/29 (45%; confidence interval [CI]: 26%–64%) kudus tested positive for FMD, but none of these reacted to SAT serotypes. For brucellosis, seropositive results were obtained for 3/29 (10%; CI: 2%–27%) kudu samples. Antibodies againstC. burnetiiwere detected in 6/29 (21%; CI: 8%–40%) kudus, 14/21 (67%; CI: 43%–85%) impalas, and 18/39 (46%; CI: 30%–63%) zebras. A total of 28/28 kudus tested positive forTheileriaspp. (100%; CI: 88%–100%) and 27/28 forAnaplasma/Ehrlichiaspp. (96%; CI: 82%–100%), whereas 12/19 impalas (63%) and 2/39 zebra (5%) tested positive forAnaplasma centrale. In ENP, only 1/29 (3%; CI: 0%–18%) wildebeest samples tested positive for FMD. None of the samples tested positive for brucellosis, whileC. burnetiiantibodies were detected in 26/30 wildebeests (87%; CI: 69%–96%), 16/40 kudus (40%; CI: 25%–57%), and 26/26 plains zebras (100%; CI: 87%–100%). A total of 60%Anaplasma/Ehrlichiaspp. and 35%Theileria/Babesiaspp. in kudu and 37% wildebeest tested positive toTheileriasp. (sable), 30% toBabesia occultans, and 3%–7% toAnaplasmaspp. The seroprevalence of Q fever was significantly higher in ENP, whileBrucellaspp.,Anaplasma,Ehrlichia,Theileria, andBabesiaspecies were significantly higher in KNP. Significant coinfections were also identified. Conclusion:This work provided baseline serological and molecular data on 40+ pathogens in four wildlife species from two national parks in southern Africa.more » « less
-
Catherino, William (Ed.)Objective: To assess whether co-culture with vitrified-warmed cumulus cells (CCs) in media drops improves rescue in vitro maturation (IVM) of previously vitrified immature oocytes. Previous studies have shown improved rescue IVM of fresh immature oocytes when cocultured with CCs in a three-dimensional matrix. However, the scheduling and workload of embryologists would benefit from a simpler IVM approach, particularly in the setting of time-sensitive oncofertility oocyte cryopreservation (OC) cases. Although the yield of developmentally competent mature metaphase II (MII) oocytes is increased when rescue IVM is performed before cryopreservation, it is unknown whether maturation of previously vitrified immature oocytes is improved after coculture with CCs in a simple system not involving a three-dimensional matrix. Design: Randomized controlled trial. Setting: Academic hospital. Patients: A total of 320 (160 germinal vesicles [GVs] and 160 metaphase I [MI]) immature oocytes and autologous CC clumps were vitrified from patients who were undergoing planned OC or intracytoplasmic sperm injection from July 2020 until September 2021. Interventions: On warming, the oocytes were randomized to culture in IVM media with CCs (+CC) or without CCs (-CC). Germinal vesicles and MI oocytes were cultured in 25 μL (SAGE IVM medium) for 32 hours and 20-22 hours, respectively. Main outcome measures: Oocytes with a polar body (MII) were randomized to confocal microscopy for analysis of spindle integrity and chromosomal alignment to assess nuclear maturity or to parthenogenetic activation to assess cytoplasmic maturity. Wilcoxon rank sum tests for continuous variables and the chi square or Fisher's exact test for categorical variables assessed statistical significance. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated. Results: Patient demographic characteristics were similar for both the GV and MI groups after randomization to +CC vs. -CC. No statistically significant differences were observed between +CC vs. -CC groups regarding the percentage of MII from either GV (42.5% [34/80] vs. 52.5% [42/80]; RR 0.81; 95% CI: 0.57-1.15]) or MI (76.3% [61/80]; vs. 72.5% [58/80]; RR 1.05; 95% CI: 0.88-1.26]) oocytes. An increased percentage of GV-matured MIIs underwent parthenogenetic activation in the +CC group (92.3% [12/13] vs. 70.8% [17/24]), but the difference was not statistically significant (RR 1.30; 95% CI: 0.97-1.75), whereas the activation rate was identical for MI-matured oocytes (74.3% [26/35] vs. 75.0% [18/24], CC+ vs. CC-; RR 0.99; 95% CI: 0.74-1.32). No significant differences were observed between +CC vs. -CC groups for cleavage of parthenotes from GV-matured oocytes (91.7% [11/12] vs. 82.4% [14/17]) or blastulation (0 for both) or for MI-matured oocytes (cleavage: 80.8% [21/26] vs. 94.4% [17/18]; blastulation: 0 [0/26] vs. 16.7% [3/18]). Further, no significant differences were observed between +CC vs. -CC for GV-matured oocytes regarding incidence of bipolar spindles (38.9% [7/18] vs. 33.3% [5/15]) or aligned chromosomes (22.2% [4/18] vs. 0.0 [0/15]); or for MI-matured oocytes (bipolar spindle: 38.9% [7/18] vs. 42.9% [2/28]); aligned chromosomes (35.3% [6/17] vs. 24.1% [7/29]). Conclusions: Cumulus cell co-culture in this simple two-dimensional system does not improve rescue IVM of vitrified, warmed immature oocytes, at least by the markers assessed here. Further work is required to assess the efficacy of this system given its potential to provide flexibility in a busy, in vitro fertilization clinic.more » « less
-
Importance Autism detection early in childhood is critical to ensure that autistic children and their families have access to early behavioral support. Early correlates of autism documented in electronic health records (EHRs) during routine care could allow passive, predictive model-based monitoring to improve the accuracy of early detection. Objective To quantify the predictive value of early autism detection models based on EHR data collected before age 1 year. Design, Setting, and Participants This retrospective diagnostic study used EHR data from children seen within the Duke University Health System before age 30 days between January 2006 and December 2020. These data were used to train and evaluate L2-regularized Cox proportional hazards models predicting later autism diagnosis based on data collected from birth up to the time of prediction (ages 30-360 days). Statistical analyses were performed between August 1, 2020, and April 1, 2022. Main Outcomes and Measures Prediction performance was quantified in terms of sensitivity, specificity, and positive predictive value (PPV) at clinically relevant model operating thresholds. Results Data from 45 080 children, including 924 (1.5%) meeting autism criteria, were included in this study. Model-based autism detection at age 30 days achieved 45.5% sensitivity and 23.0% PPV at 90.0% specificity. Detection by age 360 days achieved 59.8% sensitivity and 17.6% PPV at 81.5% specificity and 38.8% sensitivity and 31.0% PPV at 94.3% specificity. Conclusions and Relevance In this diagnostic study of an autism screening test, EHR-based autism detection achieved clinically meaningful accuracy by age 30 days, improving by age 1 year. This automated approach could be integrated with caregiver surveys to improve the accuracy of early autism screening.more » « less
-
Highly pathogenic avian influenza (HPAI) H5N1 haemagglutinin clade 2.3.4.4b was detected in the USA in 2021. These HPAI viruses caused mortality events in poultry, wild birds and wild mammals. On 25 March 2024, HPAI H5N1 clade 2.3.4.4b was confirmed in a dairy cow in Texas in response to a multistate investigation into milk production losses1. More than 200 positive herds were identified in 14 US states. The case description included reduced feed intake and rumen motility in lactating cows, decreased milk production and thick yellow milk2,3. The diagnostic investigation revealed viral RNA in milk and alveolar epithelial degeneration and necrosis and positive immunoreactivity of glandular epithelium in mammary tissue. A single transmission event, probably from birds, was followed by limited local transmission and onward horizontal transmission of H5N1 clade 2.3.4.4b genotype B3.13 (ref. 4). Here we sought to experimentally reproduce infection with genotype B3.13 in Holstein yearling heifers and lactating cows. Heifers were inoculated by an aerosol respiratory route and cows by an intramammary route. Clinical disease was mild in heifers, but infection was confirmed by virus detection, lesions and seroconversion. Clinical disease in lactating cows included decreased rumen motility, changes to milk appearance and production losses. Infection was confirmed by high levels of viral RNA detected in milk, virus isolation, lesions in mammary tissue and seroconversion. This study provides the foundation to investigate additional routes of infection, pathogenesis, transmission and intervention strategies.more » « less
