GSGustavo Madeira Santana

Gustavo Madeira Santana

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Gustavo Madeira Santana has a total of 39 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Fly navigational responses exploit plume-specific odor motion and gradient cues

    by Samuel Brudner, Baohua Zhou, Viraaj Jayaram, Gustavo Madeira Santana, John P. Crimaldi, Damon A. Clark, Thierry Emonet

    Odor cues guide animals to food and mates. Different environmental conditions can create differently patterned odor plumes, making navigation more challenging. Prior work has shown that animals turn upwind when they detect odor and cast crosswind when they lose it. Animals with bilateral olfactory sensors can also detect directional odor cues, such as odor gradient and odor motion. It remains unknown how animals use these two directional odor cues to guide crosswind navigation in odor plumes with distinct statistics. We investigated this problem theoretically and experimentally. Here, we show that these directional odor cues provide complementary information for navigation in different plume environments. We numerically analyzed experimentally generated plumes to show that odor gradient cues are more informative about crosswind directions in a smooth odor plume, while odor motion cues are more informative in a complex plume. Neural networks trained to optimize crosswind turning converged to distinctive network structures that are tuned to odor gradient cues in a smooth plume and to odor motion cues in a complex one. These trained networks improved the performance of artificial agents navigating plume environments that match the training environment. Last, we recorded Drosophila fruit flies as they navigated these odor plume environments. In the smooth plume, the gradient cue was the significant predictor of fly crosswind turning; in the complex plume, the motion cue was the significant predictor. Overall, these results demonstrate that these directional odor cues are complementary across environments and that animals exploit this relationship to navigate plumes with different statistics.


  • Open Access

    Age-specific regulation of sociability by hypothalamic Agrp neurons

    by Onur Iyilikci, Lucas Kim, Marcelo R. Zimmer, Jeremy Bober, Yuexuan Li, Macy Pelts, Gustavo M. Santana, Marcelo O. Dietrich

    Social isolation enhances sociability, suggesting that social behavior is maintained through a homeostatic mechanism. Further, mammalian social needs shift dramatically from infancy through adolescence into adulthood, raising the question of whether the neural mechanisms governing this homeostatic regulation evolve across developmental stages. Here, we show that agouti-related peptide (Agrp) neurons, which regulate hunger in adults, are activated by social isolation from weaning through adolescence but not in adulthood. Importantly, the activity of these neurons is critical for social behavior during adolescence: inhibiting Agrp neurons reduced isolation-induced sociability in juveniles but not in adults, and Agrp neuron activation promoted sociability only in young mice. After isolation, reunion with siblings or other conspecifics, but not unfamiliar adult males, rapidly decreased neuronal activity in juveniles, an effect requiring intact olfaction. These findings identify Agrp neurons as a key component of the circuitry governing age-specific social homeostasis.


  • Open Access

    Fly navigational responses to odor motion and gradient cues are tuned to plume statistics

    by Samuel Brudner, Baohua Zhou, Viraaj Jayaram, Gustavo Madeira Santana, Damon A. Clark, Thierry Emonet

    Odor cues guide animals to food and mates. Different environmental conditions can create differently patterned odor plumes, making navigation more challenging. Prior work has shown that animals turn upwind when they detect odor and cast crosswind when they lose it. Animals with bilateral olfactory sensors can also detect directional odor cues, such as odor gradient and odor motion. It remains unknown how animals use these two directional odor cues to guide crosswind navigation in odor plumes with distinct statistics. Here, we investigate this problem theoretically and experimentally. We show that these directional odor cues provide complementary information for navigation in different plume environments. We numerically analyzed real plumes to show that odor gradient cues are more informative about crosswind directions in relatively smooth odor plumes, while odor motion cues are more informative in turbulent or complex plumes. Neural networks trained to optimize crosswind turning converge to distinctive network structures that are tuned to odor gradient cues in smooth plumes and to odor motion cues in complex plumes. These trained networks improve the performance of artificial agents navigating plume environments that match the training environment. By recording Drosophila fruit flies as they navigated different odor plume environments, we verified that flies show the same correspondence between informative cues and plume types. Fly turning in the crosswind direction is correlated with odor gradients in smooth plumes and with odor motion in complex plumes. Overall, these results demonstrate that these directional odor cues are complementary across environments, and that animals exploit this relationship. Significance: Many animals use smell to find food and mates, often navigating complex odor plumes shaped by environmental conditions. While upwind movement upon odor detection is well established, less is known about how animals steer crosswind to stay in the plume. We show that directional odor cues—gradients and motion—guide crosswind navigation differently depending on plume structure. Gradients carry more information in smooth plumes, while motion dominates in turbulent ones. Neural network trained to optimize crosswind navigation reflect this distinction, developing gradient sensitivity in smooth environments and motion sensitivity in complex ones. Experimentally, fruit flies adjust their turning behavior to prioritize the most informative cue in each context. These findings likely generalize to other animals navigating similarly structured odor plumes.


  • Open Access

    SqueakOut: Autoencoder-based segmentation of mouse ultrasonic vocalizations

    by Gustavo M. Santana, Marcelo O. Dietrich

    Mice emit ultrasonic vocalizations (USVs) that are important for social communication. Despite great advancements in tools to detect USVs from audio files in the recent years, highly accurate segmentation of USVs from spectrograms (i.e., removing noise) remains a significant challenge. Here, we present a new dataset of 12,954 annotated spectrograms explicitly labeled for mouse USV segmentation. Leveraging this dataset, we developed SqueakOut , a lightweight ( 4.6M parameters) fully convolutional autoencoder that achieves high accuracy in supervised segmentation of USVs from spectrograms, with a Dice score of 90.22 . SqueakOut combines a MobileNetV2 backbone with skip connections and transposed convolutions to precisely segment USVs. Using stochastic data augmentation techniques and a hybrid loss function, SqueakOut learns robust segmentation across varying recording conditions. We evaluate SqueakOut’s performance, demonstrating substantial improvements over existing methods like VocalMat ( 63.82 Dice score). The accurate USV segmentations enabled by SqueakOut will facilitate novel methods for vocalization classification and more accurate analysis of mouse communication. To promote further research, we release the annotated 12,954 spectrogram USV segmentation dataset and the SqueakOut implementation publicly.


  • Top Cited Papers
    Open Access

    Analysis of ultrasonic vocalizations from mice using computer vision and machine learning

    by Antonio Ho Fonseca, Gustavo M Santana, Gabriela M Bosque Ortiz, Sérgio Bampi, Marcelo O Dietrich

    Mice emit ultrasonic vocalizations (USVs) that communicate socially relevant information. To detect and classify these USVs, here we describe VocalMat. VocalMat is a software that uses image-processing and differential geometry approaches to detect USVs in audio files, eliminating the need for user-defined parameters. VocalMat also uses computational vision and machine learning methods to classify USVs into distinct categories. In a data set of >4000 USVs emitted by mice, VocalMat detected over 98% of manually labeled USVs and accurately classified ≈86% of the USVs out of 11 USV categories. We then used dimensionality reduction tools to analyze the probability distribution of USV classification among different experimental groups, providing a robust method to quantify and qualify the vocal repertoire of mice. Thus, VocalMat makes it possible to perform automated, accurate, and quantitative analysis of USVs without the need for user inputs, opening the opportunity for detailed and high-throughput analysis of this behavior.


  • Exploring high-order adder compressors for power reduction in sum of absolute differences architectures for real-time UHD video encoding

    by Guilherme Paim, Gustavo M. Santana, Brunno A. Abreu, Leandro M. G. Rocha, Mateus Grellert, Eduardo A. C. da Costa, Sergio Bampi


  • Open Access

    Analysis of Ultrasonic Vocalizations from Mice Using Computer Vision and Machine Learning

    by Antonio H. O. Fonseca, Gustavo M. Santana, Sérgio Bampi, Marcelo O. Dietrich

    A bstract: Mice emit ultrasonic vocalizations (USV) to transmit socially-relevant information. To detect and classify these USVs, here we describe the development of VocalMat. VocalMat is a software that uses image-processing and differential geometry approaches to detect USVs in audio files, eliminating the need for user-defined parameter tuning. VocalMat also uses computational vision and machine learning methods to classify USVs into distinct categories. In a dataset of >4,000 USVs emitted by mice, VocalMat detected more than >98% of the USVs and accurately classified ≈86% of USVs when considering the most likely label out of 11 different USV types. We then used Diffusion Maps and Manifold Alignment to analyze the probability distribution of USV classification among different experimental groups, providing a robust method to quantify and qualify the vocal repertoire of mice. Thus, VocalMat allows accurate and highly quantitative analysis of USVs, opening the opportunity for detailed and high-throughput analysis of this behavior.


  • Open Access

    Fast Monolithic 8-2 Adder Compressor Circuit

    by Thomas Fontanari, Guilherme Paim, Leandro Mateus Giacomini Rocha, Gustavo Madeira Santana, Eduardo Antônio César Da Costa, Sergio Bampi

    Adder compressor architectures have been widely used in multipliers and have recently achieved improvements over conventional approaches in the computation of multiple modules, such as transform blocks, in the context of video coding. This paper reviews four different state-of-the-art 8-2 adder compressor architectures and proposes a novel one. A 65 nm commercial standard cell library was used to synthesize the compressors. The results show that, as a consequence of a shorter critical path, our circuit presented significant improvements regarding maximum operational frequency, while still maintaining similar results for power dissipation. Our circuit also managed to achieve a smaller circuit area, as a result of a more straightforward net interconnection.


  • Exploring approximations in 4- and 8- point DTT hardware architectures for low-power image compression

    by Guilherme Paim, Gustavo Madeira Santana, Leandro Mateus Giacomini Rocha, Leonardo Bandeira Soares, Eduardo Antônio César da Costa, Sergio Bampi


  • Power-, Area-, and Compression-Efficient Eight-Point Approximate 2-D Discrete Tchebichef Transform Hardware Design Combining Truncation Pruning and Efficient Transposition Buffers

    by Guilherme Paim, Leandro Mateus Giacomini Rocha, Gustavo Madeira Santana, Leonardo Bandeira Soares, Eduardo Antonio Cesar da Costa, Sergio Bampi

    Due to the intensive use of discrete transforms in image/video coding, the search for fast and power-efficient design approaches for their hardware implementation becomes essential. The Discrete Tchebichef Transform (DTT) represents a discrete class of the Chebyshev orthogonal polynomials, and it is an alternative for the Discrete Cosine Transform, commonly used in picture coding. The state-of-the-art approximate DTT matrix is composed of 0, 1, -1, 2, and -2 values. In this work, we propose a new approximation for the 8-point DTT, with a higher power- and compression-efficiency by exploring coefficient truncation, leading to the values 1/16, -1/16, 1/8, and -1/8. Considering operations with integers, the smaller magnitude of coefficients causes truncation in the internal transform calculations and leads to lower values for the non-diagonal residues, which reduces non-orthogonality. The results show that the proposed 8-point pruned approximate DTT hardwired ASIC solutions increase the maximum frequency up to 64%, minimize the circuit area up to 43.6%, and saves up to 65.4% in power dissipation. The results of our DTT approximation proposal mapped for FPGA show an increase of up to 58.9% on maximum frequency, and savings of about 28.7% and 32.2% on slices and dynamic power, respectively, when compared with the literature. Our approximate DTT proposal also achieves higher compression ratio and less quality loss in the compressed image, when compared to state-of-the-art approximate DTT hardware designs.