Using a multi-head, convolutional neural network with data augmentation to improve electropherogram classification performance

Taylor, Duncan

doi:10.1016/j.fsigen.2021.102605

Cited by 7 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The method used for vertebral detection was to learn the four corner points of each vertebra and a vector pointing to the centre of the vertebra. We used the data augmentation method used in previous studies to improve the accuracy of the AI algorithm [34][35][36]. In our learning process, we generated multiple variations from a single image by adding manipulations to the input image, such as black-and-white inversion, left-right inversion, and micro-noise addition.…”

Section: Learning Methods Of Presented Ai Algorithmmentioning

confidence: 99%

Comparative Accuracy of Three Artificial Intelligence Algorithms for Automatic Cobb Angle Measurement in Scoliosis: An Analysis Using Disease-specific Teaching Data

Kato,

Maeda,

Nagura

et al. 2024

Preprint

View full text Add to dashboard Cite

Spinal deformities, including adolescent idiopathic scoliosis (AIS) and adult spinal deformity (ASD), affect many patients. The measurement of the Cobb angle on coronal radiographs is essential for their diagnosis and treatment planning. To enhance the precision of Cobb angle measurements for both AIS and ASD, we developed three distinct artificial intelligence (AI) algorithms: AIS/ASD-trained AI (trained with both AIS and ASD cases); AIS-trained AI (trained solely on AIS cases); ASD-trained AI (trained solely on ASD cases). We used 1,612 whole-spine radiographs, including 1,029 AIS and 583 ASD cases with variable postures, as teaching data. We measured the major and two minor curves. To assess the accuracy, we used 285 radiographs (159 AIS and 126 ASD) as a test set and calculated the mean absolute error (MAE) and intraclass correlation coefficient (ICC) between each AI algorithm and the average of manual measurements by four spine experts. The AIS/ASD-trained AI showed the highest accuracy among the three AI algorithms. This result suggested that learning across multiple diseases rather than disease-specific training may be an efficient AI learning method. The presented AI algorithm has the potential to reduce errors in Cobb angle measurements and improve the quality of clinical practice.

show abstract

Section: Learning Methods Of Presented Ai Algorithmmentioning

confidence: 99%

Comparative Accuracy of Three Artificial Intelligence Algorithms for Automatic Cobb Angle Measurement in Scoliosis: An Analysis Using Disease-specific Teaching Data

Kato,

Maeda,

Nagura

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Efforts are underway to understand and model instrumental artifacts [ [248] , [249] , [250] , [251] ] as well as biological artifacts of the PCR amplification process such as STR stutter products [ 252 , 253 ]. Machine learning approaches are being applied to classify artifacts versus alleles with the goal to eventually replace manual data interpretation with computer algorithms [ [254] , [255] , [256] , [257] ]. One such program, FaSTR DNA, enables potential artifact peaks from stutter, pull-up, and spikes to be filtered or flagged, and a developmental validation has been published examining 3403 profiles generated with seven different STR kits [ 258 ].…”

Section: Advancements In Current Practicesmentioning

confidence: 99%

Recent advances in forensic biology and forensic DNA typing: INTERPOL review 2019–2022

Butler

2023

Forensic Science International: Synergy

View full text Add to dashboard Cite

“…Overfitting, as opposed to underfitting, is more often identified since it can be the result of small datasets that attempt to “stretch” the conclusion to previously unseen data. Overfitting can be addressed by employing large data sets (the size of which is under debate) and may also be addressed by employing data augmentation where the data sets may be artificially grown using image transformations [ 59 ]. Overfitting can also be the result of overtraining and this aspect can be addressed through cross-validation (data resampling method to assess the generalization ability of predictive models and to prevent overfitting)[ 60 ].…”

Section: Segmentationmentioning

confidence: 99%

AI-Driven Image Analysis in Central Nervous System Tumors-Traditional Machine Learning, Deep Learning and Hybrid Models

AV¹,

Zhuge²,

Zhao³

et al. 2022

J Biotechnol Biomed

View full text Add to dashboard Cite

The interpretation of imaging in medicine in general and in oncology specifically remains problematic due to several limitations which include the need to incorporate detailed clinical history, patient and disease-specific history, clinical exam features, previous and ongoing treatment, and account for the dependency on reproducible human interpretation of multiple factors with incomplete data linkage. To standardize reporting, minimize bias, expedite management, and improve outcomes, the use of Artificial Intelligence (AI) has gained significant prominence in imaging analysis. In oncology, AI methods have as a result been explored in most cancer types with ongoing progress in employing AI towards imaging for oncology treatment, assessing treatment response, and understanding and communicating prognosis. Challenges remain with limited available data sets, variability in imaging changes over time augmented by a growing heterogeneity in analysis approaches. We review the imaging analysis workflow and examine how hand-crafted features also referred to as traditional Machine Learning (ML), Deep Learning (DL) approaches, and hybrid analyses, are being employed in AI-driven imaging analysis in central nervous system tumors. ML, DL, and hybrid approaches coexist, and their combination may produce superior results although data in this space is as yet novel, and conclusions and pitfalls have yet to be fully explored. We note the growing technical complexities that may become increasingly separated from the clinic and enforce the acute need for clinician engagement to guide progress and ensure that conclusions derived from AI-driven imaging analysis reflect that same level of scrutiny lent to other avenues of clinical research.

show abstract

Using a multi-head, convolutional neural network with data augmentation to improve electropherogram classification performance

Cited by 7 publications

References 15 publications

Comparative Accuracy of Three Artificial Intelligence Algorithms for Automatic Cobb Angle Measurement in Scoliosis: An Analysis Using Disease-specific Teaching Data

Comparative Accuracy of Three Artificial Intelligence Algorithms for Automatic Cobb Angle Measurement in Scoliosis: An Analysis Using Disease-specific Teaching Data

Recent advances in forensic biology and forensic DNA typing: INTERPOL review 2019–2022

AI-Driven Image Analysis in Central Nervous System Tumors-Traditional Machine Learning, Deep Learning and Hybrid Models

Contact Info

Product

Resources

About