Yazar "Danda, Sai Santosh Reddy" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • [ X ]
    Öğe
    Clinical Manifestations
    (2025) Uysal, Alper Kürşat; Danda, Sai Santosh Reddy; Cook Maher, Amanda; Persad, Carol C.; Rose, Savannah G.; Koeppe, Robert A.; Giordani, Bruno J.
    BACKGROUND: Identifying subtle changes in cognitively demanding tasks for older persons, such as daily driving, may reveal connections to early signs of cognitive decline associated with Alzheimer's disease (AD). Research has shown that brain amyloid positivity is strongly associated with AD. This study analyzes various trip-based quantitative measures derived from naturalistic driving to distinguish older adults with amyloid-positivity (A?+) from those who are amyloid-negative (A?-). Significant trip-based naturalist driving attributes identified in this study will be used in future development of machine learning models for classifying A?+ and A?- participants. METHOD: We analyzed naturalistic driving data from 30 A?+ (2,577 trips) and 30 A?- (2,533 trips) consensus diagnosed participants. An internet facilitated data acquisition system was installed in each participant's vehicle, and driving trips were recorded over approximately one month between 2021 and 2024. The mean ages of the A?+ and A?- groups were 72.9±4.2 and 71.4±5.6 respectively. Each recorded trip included three types of data-videos, vehicle signals, and physiological signals. We studied 22 trip-based attributes, some of which had not been examined previously in the research community, including attributes derived from valid and invalid trips. A valid trip contains all required data signals, while an invalid trip has one or more signals missing. Statistical t-tests (?=0.05) were used to evaluate the quantitative attributes of the two groups. RESULT: The following attributes were found to be significant: A?- drivers had a higher percentage of valid trips (p = 0.026), while A?+ drivers had longer average trip duration (p = 0.037), longer average trip distance (p = 0.049), and a higher percentage of weekday night trips (p = 0.048). CONCLUSION: Our research demonstrated that A?- participants had higher percentage of valid trips, potentially indicating better capabilities to operate the data acquisition devices. The longer average trip durations and distances observed in A?+ participants may reflect less efficient route choices while driving. Additionally, the higher percentage of weekday night trips among A?+ participants suggests higher potential risk of more nighttime driving activity. © 2025 The Alzheimer's Association. Alzheimer's & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer's Association.
  • [ X ]
    Öğe
    Deep Neural Network Ensembles for the Detection of Alzheimer's Disease Using Imbalanced Clock Drawing Test Images
    (Institute of Electrical and Electronics Engineers Inc., 2025) Danda, Sai Santosh Reddy; Uysal, Alper Kürşat; Qin, Tian; Kumar, Anu M.; Sannareddy, Varshitha; Hu, Mengyao; Gonzalez, Richard D.
    This paper explores innovative machine learning technologies for the detection of Alzheimer's Disease (AD) from large and imbalanced clock drawing test (CDT) images. We present a deep ensemble learning framework, 2Level MM-MD, which includes two-level ensemble of MultiModel deep neural networks (DNN) trained on MultiData blocks. The data blocks are generated by applying partition based and bootstrap statistical data resampling methods to the large but imbalanced CDT image data. To produce a more robust system that is better adapted to imbalanced CDT data, the 2Level MM-MD ensemble framework attempts to leverages both 1) the richness of large data blocks generated using effective data re-sampling schemes, and 2) multiple deep neural network systems that are trained on the resembled data blocks through transfer learning to produce a more robust system that is better adapted to more generalized CDT image domains. Three deep neural network models (EfficientNet, ResNet101, and ViT), along with six different sampling methods are used to generate re-balanced data blocks, and five different decision schemes are implemented and evaluated. The ensemble systems are trained and evaluated using a large CDT image dataset from Rounds 1 to 9 of the National Health and Aging Trends Study (NHATS), which contains over 47,723 CDT images, where 92% are class 0 (no dementia) and 8% are class 1 (probable dementia). The top classification system is a 2Level MM-MD system E5 with a two-level decision rule SoM. The system generated 0.67 Recall and 0.35 F1-Score on the test data. This represents a 123% improvement in Recall and a 25% improvement in F1-Score compared to the best baseline system, which is a classifier trained using EfficientNet with the training data balanced by oversampling the minority class. © 2025 IEEE.