Machine Learning Methods for Small Data Challenges in Molecular Science

Chemical Reviews, Journal Year: 2023, Volume and Issue: 123(13), P. 8736 - 8780

Published: June 29, 2023

Small data are often used in scientific and engineering research due to the presence of various constraints, such as time, cost, ethics, privacy, security, technical limitations acquisition. However, big have been focus for past decade, small their challenges received little attention, even though they technically more severe machine learning (ML) deep (DL) studies. Overall, challenge is compounded by issues, diversity, imputation, noise, imbalance, high-dimensionality. Fortunately, current era characterized technological breakthroughs ML, DL, artificial intelligence (AI), which enable data-driven discovery, many advanced ML DL technologies developed inadvertently provided solutions problems. As a result, significant progress has made decade. In this review, we summarize analyze several emerging potential molecular science, including chemical biological sciences. We review both basic algorithms, linear regression, logistic regression (LR),

Language: Английский

---

The role of population structure in computations through neural dynamics

Nature Neuroscience, Journal Year: 2022, Volume and Issue: 25(6), P. 783 - 794

Published: June 1, 2022

Language: Английский

---

Interpreting neural computations by examining intrinsic and embedding dimensionality of neural activity

Current Opinion in Neurobiology, Journal Year: 2021, Volume and Issue: 70, P. 113 - 120

Published: Sept. 17, 2021

Language: Английский

---

The neuroconnectionist research programme

Nature reviews. Neuroscience, Journal Year: 2023, Volume and Issue: 24(7), P. 431 - 450

Published: May 30, 2023

Language: Английский

---

Memristor-based neural networks: a bridge from device to artificial intelligence

Nanoscale Horizons, Journal Year: 2023, Volume and Issue: 8(6), P. 716 - 745

Published: Jan. 1, 2023

This paper reviews the research progress in memristor-based neural networks and puts forward future development trends.

Language: Английский

---

Class imbalance should not throw you off balance: Choosing the right classifiers and performance metrics for brain decoding with imbalanced data

NeuroImage, Journal Year: 2023, Volume and Issue: 277, P. 120253 - 120253

Published: June 28, 2023

Machine learning (ML) is increasingly used in cognitive, computational and clinical neuroscience. The reliable efficient application of ML requires a sound understanding its subtleties limitations. Training models on datasets with imbalanced classes particularly common problem, it can have severe consequences if not adequately addressed. With the neuroscience user mind, this paper provides didactic assessment class imbalance problem illustrates impact through systematic manipulation data ratios (i) simulated (ii) brain recorded electroencephalography (EEG), magnetoencephalography (MEG) functional magnetic resonance imaging (fMRI). Our results illustrate how widely-used Accuracy (Acc) metric, which measures overall proportion successful predictions, yields misleadingly high performances, as increases. Because Acc weights per-class correct predictions proportionally to size, largely disregards performance minority class. A binary classification model that learns systematically vote for majority will yield an artificially decoding accuracy directly reflects between two classes, rather than any genuine generalizable ability discriminate them. We show other evaluation metrics such Area Under Curve (AUC) Receiver Operating Characteristic (ROC), less Balanced (BAcc) metric - defined arithmetic mean sensitivity specificity, provide more evaluations data. findings also highlight robustness Random Forest (RF), benefits using stratified cross-validation hyperprameter optimization tackle imbalance. Critically, applications seek minimize error, we recommend routine use BAcc, specific case balanced equivalent standard Acc, readily extends multi-class settings. Importantly, present list recommendations dealing data, well open-source code allow community replicate extend our observations explore alternative approaches coping

Language: Английский

---

Exploring data mining and machine learning in gynecologic oncology

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 57(2)

Published: Jan. 29, 2024

Abstract Gynecologic (GYN) malignancies are gaining new and much-needed attention, perpetually fueling literature. Intra-/inter-tumor heterogeneity “frightened” global distribution by race, ethnicity, human development index, pivotal clues to such ubiquitous interest. To advance “precision medicine” downplay the heavy burden, data mining (DM) is timely in clinical GYN oncology. No consolidated work has been conducted examine depth breadth of DM applicability as an adjunct oncology, emphasizing machine learning (ML)-based schemes. This systematic literature review (SLR) synthesizes evidence fill knowledge gaps, flaws, limitations. We report this SLR compliance with Kitchenham Charters’ guidelines. Defined research questions PICO crafted a search string across five libraries: PubMed, IEEE Xplore, ScienceDirect, SpringerLink, Google Scholar—over past decade. Of 3499 potential records, 181 primary studies were eligible for in-depth analysis. A spike (60.53%) corollary cervical neoplasms denoted onward 2019, predominantly featuring empirical solution proposals drawn from cohorts. Medical records led (23.77%, 53 art.). DM-ML use primarily built on neural networks (127 art.), appoint classification (73.19%, 172 art.) diagnoses (42%, 111 all devoted assessment. Summarized sufficient guide support utility schemes Gaps persist, inculpating interoperability single-institute scrutiny. Cross-cohort generalizability needed establish while avoiding outcome reporting bias locally, site-specific trained models. exempt ethics approval it entails published articles.

Language: Английский

---

Artificial Intelligence and Neuroscience: Transformative Synergies in Brain Research and Clinical Applications

Journal of Clinical Medicine, Journal Year: 2025, Volume and Issue: 14(2), P. 550 - 550

Published: Jan. 16, 2025

The convergence of Artificial Intelligence (AI) and neuroscience is redefining our understanding the brain, unlocking new possibilities in research, diagnosis, therapy. This review explores how AI’s cutting-edge algorithms—ranging from deep learning to neuromorphic computing—are revolutionizing by enabling analysis complex neural datasets, neuroimaging electrophysiology genomic profiling. These advancements are transforming early detection neurological disorders, enhancing brain–computer interfaces, driving personalized medicine, paving way for more precise adaptive treatments. Beyond applications, itself has inspired AI innovations, with architectures brain-like processes shaping advances algorithms explainable models. bidirectional exchange fueled breakthroughs such as dynamic connectivity mapping, real-time decoding, closed-loop systems that adaptively respond states. However, challenges persist, including issues data integration, ethical considerations, “black-box” nature many systems, underscoring need transparent, equitable, interdisciplinary approaches. By synthesizing latest identifying future opportunities, this charts a path forward integration neuroscience. From harnessing multimodal cognitive augmentation, fusion these fields not just brain science, it reimagining human potential. partnership promises where mysteries unlocked, offering unprecedented healthcare, technology, beyond.

Language: Английский

---

50 years of mnemonic persistent activity: quo vadis?

Trends in Neurosciences, Journal Year: 2021, Volume and Issue: 44(11), P. 888 - 902

Published: Oct. 13, 2021

Language: Английский

---

The Best Laid Plans: Computational Principles of Anterior Cingulate Cortex

Trends in Cognitive Sciences, Journal Year: 2021, Volume and Issue: 25(4), P. 316 - 329

Published: Feb. 15, 2021

Language: Английский

---