TRAFIC: fiber tract classification using deep learning

Lam, Prince D. Ngattai; Belhomme, Gaetan; Ferrall, Jessica; Patterson, Billie; Styner, Martin; Prieto, Juan-Carlos

doi:10.1117/12.2293931

Cited by 15 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One tool that is currently being tested utilizes a smartphone app and associated machine learning algorithm to identify TT in photographs. 19 Such tools could modify the differences between house-to-house and community mobilization approaches.…”

Section: Discussionmentioning

confidence: 99%

Effectiveness of Trachomatous Trichiasis Case-identification Approaches in Ethiopia

Diallo,

Bayissasse,

Sisay

et al. 2023

Epidemiology

View full text Add to dashboard Cite

Background: Trachoma control programs use multiple approaches to identify individuals with trachomatous trichiasis (TT). Evidence is limited regarding which approaches are most effective and cost-efficient. Methods: We evaluated the effectiveness of two TT case-identification approaches in Ethiopia: community mobilization to encourage self-referral for centralized screening and house-to-house screenings conducted by case finders. We compared the number of true cases found per 1000 population and costs associated with case identification under each approach, stratified by villages that received one or multiple screening visits. Results: We conducted screenings in 396 villages. In villages receiving one house-to-house visit, case finders identified 14,229 suspected cases, of whom 10,513 (73.9%) presented for TT confirmation. A median of 17.2% (interquartile range [IQR]: 9.1%–27.8%) of those presenting truly had TT (positive predictive value). In single-visit villages, the community mobilization approach yielded higher rates of confirmed cases than the house-to-house approach (1.5 [IQR: 1.1, 2.6] vs. 1.1 [IQR: 0.5, 1.9] cases per 1000 population), and the median cost of identifying a TT case was less ($5.59 vs. $31.18) using community mobilization than house-to-house. In multiple-visit villages, additional screening visits increased the median rate of confirmed cases to 2.5 per 1000 population in community mobilization villages, but the rate remained unchanged in house-to-house villages. Conclusions: Community mobilization-based TT case finding had a higher yield than house-to-house, at a substantially lower cost. Future research should examine whether additional tools to aid case finders in their diagnosis increases case-finding efficiency and accuracy and whether TT prevalence and surgical program duration impact case-finding success.

show abstract

Section: Discussionmentioning

confidence: 99%

Effectiveness of Trachomatous Trichiasis Case-identification Approaches in Ethiopia

Diallo,

Bayissasse,

Sisay

et al. 2023

Epidemiology

View full text Add to dashboard Cite

show abstract

“…Nonlinear, diffeomorphic pair-wise registration was performed to map individual subject DTIs into atlas space, and registration accuracy was visually inspected in DTI-AtlasBuilder to determine if the computed transforms were appropriate. Major fiber tracts were determined semi-automatically in this atlas space (Ngattai Lam et al, 2018). Resulting deformation fields were then used to map the atlas fibers into individual subject space, where diffusion tensor metrics were extracted at evenly spaced points (arc lengths) along each fiber tract.…”

Section: Image Processingmentioning

confidence: 99%

Exposure to prenatal maternal distress and infant white matter neurodevelopment

et al. 2021

Self Cite

View full text Add to dashboard Cite

The prenatal period represents a critical time for brain growth and development. These rapid neurological advances render the fetus susceptible to various influences with life-long implications for mental health. Maternal distress signals are a dominant early life influence, contributing to birth outcomes and risk for offspring psychopathology. This prospective longitudinal study evaluated the association between prenatal maternal distress and infant white matter microstructure. Participants included a racially and socioeconomically diverse sample of 85 mother–infant dyads. Prenatal distress was assessed at 17 and 29 weeks’ gestational age (GA). Infant structural data were collected via diffusion tensor imaging (DTI) at 42–45 weeks’ postconceptional age. Findings demonstrated that higher prenatal maternal distress at 29 weeks’ GA was associated with increased fractional anisotropy, b = .283, t(64) = 2.319, p = .024, and with increased axial diffusivity, b = .254, t(64) = 2.067, p = .043, within the right anterior cingulate white matter tract. No other significant associations were found with prenatal distress exposure and tract fractional anisotropy or axial diffusivity at 29 weeks’ GA, or earlier in gestation.

show abstract

“…Apart from the named end-to-end approaches, there are also works that only replace parts of the classical processing pipeline for dMRI by deep learning. Some methods replace the computation of diffusion tensor images (Tian et al, 2020;Li et al, 2020) or neural fibers (Nath et al, 2019b) from dMRI scans, while others use classically computed diffusion tensor images (Marzban et al, 2020) or neural fibers (Prieto et al, 2018) as inputs. These methods only replace parts of the processing pipeline for dMRI data, whereas the present work replaces the whole pipeline end-to-end, which proves more optimal for dMRI (Golkov et al, 2016a) and is the reason for the success of deep learning in general.…”

Section: Related Workmentioning

confidence: 99%

Rotation-Equivariant Deep Learning for Diffusion MRI

Müller,

Golkov,

Tomassini

et al. 2021

Preprint

View full text Add to dashboard Cite

Convolutional networks are successful, but they have recently been outperformed by new neural networks that are equivariant under rotations and translations. These new networks work better because they do not struggle with learning each possible orientation of each image feature separately. So far, they have been proposed for 2D and 3D data. Here we generalize them to 6D diffusion MRI data, ensuring joint equivariance under 3D roto-translations in image space and the matching 3D rotations in q-space, as dictated by the image formation. Such equivariant deep learning is appropriate for diffusion MRI, because microstructural and macrostructural features such as neural fibers can appear at many different orientations, and because even non-rotation-equivariant deep learning has so far been the best method for many diffusion MRI tasks. We validate our equivariant method on multiple-sclerosis lesion segmentation. Our proposed neural networks yield better results and require fewer scans for training compared to non-rotation-equivariant deep learning. They also inherit all the advantages of deep learning over classical diffusion MRI methods. Our implementation is available at https://github. com/philip-mueller/equivariant-deep-dmri and can be used off the shelf without understanding the mathematical background.

show abstract

TRAFIC: fiber tract classification using deep learning

Cited by 15 publications

References 11 publications

Effectiveness of Trachomatous Trichiasis Case-identification Approaches in Ethiopia

Effectiveness of Trachomatous Trichiasis Case-identification Approaches in Ethiopia

Exposure to prenatal maternal distress and infant white matter neurodevelopment

Rotation-Equivariant Deep Learning for Diffusion MRI

Contact Info

Product

Resources

About