Skip to content

Let's Data ScienceLEARN • BUILD • STAY AHEAD

News
Blog
Code Problems
Pricing
Contact

© 2026 Let's Data Science

Advertise|Terms|Privacy||Image Rights

Home
News
Machine Learning

Topic10 articles · Updated June 25, 2026

Machine Learning News

Machine learning coverage for working practitioners: research papers worth reading, framework and library updates, MLOps tooling, and applied ML stories shipping to production.

quadsci
4.0
Jun 25, 2026
QuadSci Wins Machine Learning Company of the Year Award
AI Breakthrough's 9th annual AI Breakthrough Awards named QuadSci the 2026 "Machine Learning Company of the Year," according to a Business Wire press release distributed on June 25, 2026. The announcement says this is the second consecutive year QuadSci has earned the award (Business Wire). The company describes its platform as ingesting product telemetry, CRM, and conversational data and delivering 90% predictive accuracy for churn and growth events 9-18 months ahead of renewal, and reports it has analyzed over 11 trillion telemetry events (Business Wire). QuadSci also reports it finds, on average, 15% of customers' ARR in a predictive growth state with no open pipeline and has integrations with more than 50 platforms including Salesloft and Salesforce (Business Wire).
Quick Read+2
systematic-review
5.7
Jun 23, 2026
Systematic Review Examines ML Prognostic Models for Spinal Cord Injury
This systematic review examines use of machine learning to develop prognostic prediction models for functional outcomes in spinal cord injury. The review reports that current studies exhibit heterogeneity and presents a systematic assessment of the literature.
Quick Read
electronic-nose
6.2
Jun 19, 2026
UC Berkeley Builds Electronic Nose for Food Safety
Researchers at UC Berkeley have developed a 16-element "electronic nose" chip that combines miniature gas sensors and machine learning to identify food types, spoilage, and common allergens, according to a paper published in Science Advances and reporting by Berkeley News. The device uses different gas-sensitive coatings on each sensor to produce a collective "fingerprint," and the team reports sensitivity down to 0.05 grams of walnut material in lab tests, per Hackster.io and Yahoo Tech coverage of the study. The chip uses carbon-nanotube sensor layers and operates at room temperature, which the research coverage says simplifies manufacturing and lowers power requirements. Editorial analysis: Industry observers should view this as an early but practical demonstration of combining multiplexed gas sensing with ML for consumer-facing food-safety applications.
Quick Read+10
actigraphy
6.1
Jun 18, 2026
ActiTect delivers generalizable RBD screening via actigraphy
According to the arXiv preprint (arXiv:2511.05221) and an online-ahead-of-print article in NPJ Digital Medicine, the authors present ActiTect, an open-source, fully automated machine learning pipeline to detect REM sleep behavior disorder (RBD) from wrist actigraphy. The paper reports model development on a cohort of 78 individuals with nested cross-validation achieving AUROC = 0.95 (arXiv). Generalization was tested on a blinded local test set (n = 31, AUROC = 0.86) and two external cohorts (n = 113, AUROC = 0.84; n = 57, AUROC = 0.94), per NPJ Digital Medicine and the arXiv record. The study also reports leave-one-dataset-out cross-validation with an AUROC range = 0.84-0.89 and a stability analysis showing reproducible predictive features across datasets. The authors state the code is open-source to encourage independent validation and collaborative improvement (arXiv; NPJ Digital Medicine).
Quick Read
lyme-disease
6.9
Jun 17, 2026
ML Identifies Borrelia burgdorferi Determinants of Dissemination
Per a study in PLOS Computational Biology (Nguyen and Brissette, originating from a July 2025 bioRxiv preprint), researchers applied machine learning to whole-genome sequences from 299 clinical isolates of Borrelia burgdorferi to link protein sequence variation with human dissemination phenotypes. The study extracted variants of seven known virulence factors, including BB_0406, BBK32, DbpA, OspA, OspC, P66, and RevA, and used Cramer's V to find strong associations between dissemination and five adhesins: BBK32, DbpA, OspC, P66, and RevA. The authors trained models with five algorithms and multiple feature-selection strategies, reporting predictive performance above 0.7 for DbpA, OspC, and RevA variants. Feature-importance analysis highlighted specific amino-acid residues, and B-cell epitope prediction showed enrichment of ML-identified residues for OspC and RevA.
Quick Read+1
cardiology
5.6
Jun 9, 2026
Paper presents Cardiology Interface Terminology curated with machine learning
According to the arXiv paper, the authors propose a Cardiology Interface Terminology (CIT) and a three-phase workflow that combines semi-automatic phrase mining and a machine learning stage to curate the terminology for highlighting cardiology electronic health record (EHR) notes. The arXiv submission reports that the highlighted test set achieves a coverage of 74.21% and a breadth of 1.68, and that for 20 random test notes average completeness is 98.2% and conciseness 84.2% (arXiv:2606.08311). A separate ResearchWithNJ conference chapter reports an ML-assisted curation variant (CIT ML2+) whose test coverage is 68.74%, with near-equivalent completeness and conciseness compared to a fully manually curated CIT+ (ResearchWithNJ). Editorial analysis: For clinical-NLP practitioners, the paper documents a pragmatic path from SNOMED CT seed concepts to a higher-granularity, domain-specific interface terminology and shows measurable tradeoffs between manual review effort and coverage when using an ML-assisted curation loop.
Quick Read+6
quantum-photonics
5.5
Jun 8, 2026
Authors optimize two-color nanofiber trap with machine learning
An arXiv preprint (arXiv:2606.06798) reports an experimental realization of a two-color dipole trap around an optical nanofiber and the use of a machine-learning algorithm to maximise the trapped-atom number. According to the preprint, the authors estimate roughly 1400 atoms in the trap and a trap lifetime of 28 ms. The paper reports that machine-learning optimisation increased the measured on-resonance optical depth from 0.5 in an initial stage to values exceeding 15, with optical density inferred from transmission measurements. The work was submitted to arXiv on June 5, 2026, by researchers at the University of Auckland.
Quick Read
wildlife-trafficking
6.3
Jun 8, 2026
AI Detects Smuggled Seahorses and Shark Fins
Researchers trained an AI algorithm on hundreds of three-dimensional x-ray scans of 68 dried marine samples to detect smuggled seahorses, shark fins and sea cucumbers, reporting an overall detection accuracy of 92% and a false positive rate near 13%, according to the study published in Frontiers in Ocean Sustainability and reported by Scientific American, The Conversation and Phys.org. Per the Frontiers article, subclass performance was 95% for shark fins, 96% for seahorses and 86% for sea cucumbers. Lead author Vanessa Pirotta, a wildlife scientist at Macquarie University, is quoted in multiple outlets on the potential to deploy the approach at airports; The Conversation discloses Pirotta received funding from Rapiscan Systems for the research.
Quick Read+1
gastrointestinal-cancer
5.6
Jun 4, 2026
Study Advances GI Cancer Risk Prediction with ML
A peer-reviewed study published in JMIR Medical Informatics evaluates machine learning methods for predicting gastrointestinal (GI) cancer risk in a South Korean prospective cohort, where GI cancers are a major health burden. Analyzing 7,652 participants with 156 incident GI cancer cases (about 2%) over 14 years of follow-up, the authors tackle the severe class imbalance that makes rare-disease prediction difficult. They introduce a patient-centered undersampling technique (PCUSTe) modeled on frequency-matched case-control design and benchmark it against SMOTE, ADASYN, and hybrid resampling. An incrementally trained stochastic gradient descent model on PCUSTe data reached a sensitivity of 0.77 and an AUC of 0.77, while logistic regression without resampling produced balanced results (sensitivity 0.70, specificity 0.71, AUC 0.75). The authors frame the models as tools for earlier risk stratification and targeted screening.
Quick Read
wearables
6.5
Jun 3, 2026
Researchers Develop Compact Wearable for Continuous Blood Pressure
A team from the University of Utah and the University of Illinois Chicago, with collaborators at Harvard Medical School and the University of Pittsburgh, built a cuffless, smartwatch-style wearable that measures blood pressure continuously, the University of Utah reports. The device passes an imperceptible electrical current through the wrist and uses bioimpedance - how easily current flows through blood and tissue - to capture changes with each heartbeat, then reconstructs a full blood-pressure waveform with a physics-informed machine learning model. Embedding the physics of blood flow keeps the model from predicting physically impossible values and, the team says, removes the need for per-user calibration. Graduate students tested it on 150 people, including intensive-care and outpatient patients. The work is set to appear in Nature Communications as Cuffless hemodynamic monitoring with physics-informed machine learning models, with an early version posted May 14.
Quick Read+1

See the full news feed