Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (3 results)

  • Thesis
  • Peer Reviewed
UC Santa Cruz Electronic Theses and Dissertations (2012)

Sea otters (Enhydra lutris) are well known and conspicuous tool users, but little is known about what drives the maintenance of this behavior in populations or individuals. I investigated how variation in the frequency of tool use across and within sea otter populations may be influenced by ecological factors such as age class, sex and reproductive status, geographic location, feeding habitat, and prey type. Additionally, I explored whether consistent inter-individual differences in tool use occurred and if the frequency of tool use occurrence was related to learned diet specializations. I used observed foraging and tool use data collected from nine sites across two subspecies of sea otters. Over 500 individuals were observed feeding and over 100,000 feeding dives were recorded between 1985 and 2011. Using binary generalized linear mixed effects models, I found the type of prey consumed the strongest predictor of the frequency of tool use across populations, although all factors contributed to the best fit model. In Monterey, California, I collected longitudinal data on sixty- three individuals for a minimum of one year. I found that individuals specializing in prey that required tools were more likely to carry-over this behavior to other prey consumed. My results suggest that the frequency of tool use in sea otter populations and individuals is influenced by ecological factors such as the consumption of prey that is difficult to access, and the "behavioral inertia" of individuals learning to use tools for particular prey items.

    Cover page: Variation in the frequency of tool use across and within sea otter (Enhydra lutris) populations
    • Article
    • Peer Reviewed
    UC Davis Previously Published Works (2023)

    Rationale and objectives

    Spinal osteoporotic compression fractures (OCFs) can be an early biomarker for osteoporosis but are often subtle, incidental, and underreported. To ensure early diagnosis and treatment of osteoporosis, we aimed to build a deep learning vertebral body classifier for OCFs as a critical component of our future automated opportunistic screening tool.

    Materials and methods

    We retrospectively assembled a local dataset, including 1790 subjects and 15,050 vertebral bodies (thoracic and lumbar). Each vertebral body was annotated using an adaption of the modified-2 algorithm-based qualitative criteria. The Osteoporotic Fractures in Men (MrOS) Study dataset provided thoracic and lumbar spine radiographs of 5994 men from six clinical centers. Using both datasets, five deep learning algorithms were trained to classify each individual vertebral body of the spine radiographs. Classification performance was compared for these models using multiple metrics, including the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, and positive predictive value (PPV).

    Results

    Our best model, built with ensemble averaging, achieved an AUC-ROC of 0.948 and 0.936 on the local dataset's test set and the MrOS dataset's test set, respectively. After setting the cutoff threshold to prioritize PPV, this model achieved a sensitivity of 54.5% and 47.8%, a specificity of 99.7% and 99.6%, and a PPV of 89.8% and 94.8%.

    Conclusion

    Our model achieved an AUC-ROC>0.90 on both datasets. This testing shows some generalizability to real-world clinical datasets and a suitable performance for a future opportunistic osteoporosis screening tool.
      Cover page: Generalizability of Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs Using an Adaptation of the Modified-2 Algorithm-Based Qualitative CriteriaCreative Commons 'BY' version 4.0 license
      • Article
      • Peer Reviewed
      UC Davis Previously Published Works (2022)

      Rationale and objectives

      Osteoporosis affects 9% of individuals over 50 in the United States and 200 million women globally. Spinal osteoporotic compression fractures (OCFs), an osteoporosis biomarker, are often incidental and under-reported. Accurate automated opportunistic OCF screening can increase the diagnosis rate and ensure adequate treatment. We aimed to develop a deep learning classifier for OCFs, a critical component of our future automated opportunistic screening tool.

      Materials and methods

      The dataset from the Osteoporotic Fractures in Men Study comprised 4461 subjects and 15,524 spine radiographs. This dataset was split by subject: 76.5% training, 8.5% validation, and 15% testing. From the radiographs, 100,409 vertebral bodies were extracted, each assigned one of two labels adapted from the Genant semiquantitative system: moderate to severe fracture vs. normal/trace/mild fracture. GoogLeNet, a deep learning model, was trained to classify the vertebral bodies. The classification threshold on the predicted probability of OCF outputted by GoogLeNet was set to prioritize the positive predictive value (PPV) while balancing it with the sensitivity. Vertebral bodies with the top 0.75% predicted probabilities were classified as moderate to severe fracture.

      Results

      Our model yielded a sensitivity of 59.8%, a PPV of 91.2%, and an F1 score of 0.72. The areas under the receiver operating characteristic curve (AUC-ROC) and the precision-recall curve were 0.99 and 0.82, respectively.

      Conclusion

      Our model classified vertebral bodies with an AUC-ROC of 0.99, providing a critical component for our future automated opportunistic screening tool. This could lead to earlier detection and treatment of OCFs.
        Cover page: Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs using an Adaptation of the Genant Semiquantitative Criteria
        Top