hextrato

SYNNER (Synthetic Data) in 3 phases (and why that matters) A big lesson from building in sensitive domains: you can’t treat utility and privacy as an afterthought. SYNNER is structured into three phases: Read more (paper): journals.sagepub.com/doi/10.1177/20…

FormulAI: Designing Rule-Based Datasets for Interpretable and Challenging Machine Learning Tasks link.growkudos.com/1bgqzck2vi8

Improving Risk Assessment of Miscarriage During Pregnancy with Knowledge Graph Embeddings link.growkudos.com/1egf4zxrugw

We'd love to hear your thoughts! How might the discovery of baseline models matching deep learning impact healthcare practices? Share your views! #HealthcareDiscussion

Do you think a simple Baseline ML Model can defeat a more sophisticated approach? Can a baseline be David against the Deep Learning Goliath? Check our findings: ojs.luminescience.cn/JDH/article/vi… #MachineLearning #BaselineModels #DeepLearning #ICDCoding #HealthInformatics

Our study reveals that baseline models can match the efficiency of deep learning in ICD coding. How might this reshape healthcare practices?

What if traditional methods perform as well as sophisticated algorithms? Intrigued?

Check what we found in our recent publication about ICD Coding from discharge notes. (see link in main post) #ICD #ICDCoding #DischargeNotes #HealthInformatics

Baseline vs Deep Learning In the healthcare context, when baseline models do not just serve to establish a minimum capacity for performance, they might be comparable to other more sophisticated deep learning approaches. ojs.luminescience.cn/JDH/article/vi…

A thorough evaluation of different models and approaches is crucial to avoid over-reliance on any method. Our findings show that simpler methods can achieve comparable results to deep learning models while still requiring fewer computational resources.

ICD Coding Our paper comparing NLP and baseline models is available online now: hextrato.com/MyPublications… #ICDCoding #DataAnalysis #DataDrivenHealthcare #HealthcareInnovation #MachineLarning

We investigated automatic ICD coding using baseline machine learning models, with a focus on identifying ICD-9 codes in discharge notes from Medical Information Mart for Intensive Care (MIMIC) database.

Evaluating deep learning models under the scenarios of real-world coding and comparing them to established practice is critical to determining their true effectiveness.

Although deep learning shows promise in healthcare, its specific impact on ICD coding is not fully understood, and translating scalable deep learning methods into practical improvements in ICD coding remains a challenge.

Accurate coding requires a deep understanding of medical terminology, context, and guidelines that may be difficult to capture with traditional deep learning methods.

To simplify the process, clinical records are often processed to automatically identify and extract clinical concepts and corresponding ICD codes.

Manual medical coding is a labor-intensive and error-prone process that creates additional administrative burden and inconvenience for hospitals, insurance companies, and patients.

Incorrect codes waste time and resources, and cause administrative and financial problems for hospitals, insurance companies and patients.

Accuracy in assigning International Classification of Diseases (ICD) codes is critical to proper reimbursement of care.

Billing involves filing claims with insurance companies and requires scrutiny of clinical summaries and electronic health records to correctly match diagnoses, prescriptions, and procedures to standardized codes.

Financial costs are a major concern in the healthcare system, with medical billing and coding playing a key role in facilitating transactions and financing procedures.