Skip to content
Calibrated Researcher

Professor of computer science, Hassan First University of Settat

Confidence that matches the evidence. In your models and in your papers.

I am Ilhame Ait Lbachir, a computer science professor and AI researcher. I help researchers do work that holds up to scrutiny: stronger research, sharper papers, proposals, and talks, with a specialist’s eye for trustworthy, uncertainty-aware AI.

Two mammogram views of one lesion stacked in 3D, with a surface rising over the lesion where the model is least certain
Two mammogram views of one lesion, from my research data. The surface sketches where a model’s uncertainty concentrates.
10 yearsin research
About 50peer reviews
IEEE and Springerpublished
IWBBIO 2026conference talk

In the room

Presenting my work on uncertainty in breast cancer imaging, to researchers and clinicians.

Ilhame Ait Lbachir presenting a comparison of uncertainty quantification reviews for breast cancer computer-aided diagnosis
Presenting a review of uncertainty quantification for breast cancer CAD.
Ilhame Ait Lbachir speaking at the podium, with the event banner of the AI day organised by the FST Settat
AI Day, Settat.

Help you do work that holds up to scrutiny

Not louder work. Stronger work.

01

Research that holds up to scrutiny

Methodology and experimental design that survive a hard read. How to frame a question, run experiments that mean something, and report results reviewers trust.

02

Papers, proposals, and talks that get accepted

The craft of writing for acceptance. How a paper actually gets written, where proposals fall apart, and what reviewers look for before they say yes.

03

A specialist’s lens on trustworthy AI

From my own research: uncertainty quantification, explainability, and calibration for medical imaging. How to reason about AI you can actually trust, made clear.

The Calibrated Method

The standard I hold myself to, and the one I teach: confidence that matches the evidence, never more.

01

Claim only what the evidence supports

Every sentence earns its place. If the data does not back it, it does not ship.

02

Stress-test before submission

Read your own work the way a hostile reviewer will. Find the holes before they do.

03

Report uncertainty honestly

Say what you do not know. Calibrated confidence is more persuasive than false certainty.

Uncertainty Quantification for Trustworthy Breast Cancer CAD in Mammography: A Critical Review and Research Agenda

I. Ait Lbachir, I. Daoudi, R. Nassih · International Conference on Intelligent Systems and Digital Applications (ISDA), IEEE

Published, 2026

Digital Breast Tomosynthesis Reconstruction Techniques in Healthcare Systems: A Review

I. Samiry, I. Ait Lbachir, I. Daoudi, S. Tallal, S. Adil · International Work-Conference on Bioinformatics and Biomedical Engineering (IWBBIO), Springer

Published, 2023

Automatic computer-aided diagnosis system for mass detection and classification in mammography

I. Ait Lbachir, I. Daoudi, S. Tallal · Multimedia Tools and Applications (vol. 80), Springer

Journal, 2021

Field notes

New here? Start with the field notes.

All field notes

The pre-submission checklist reviewers wish you used.

The checks I run before any paper goes out, distilled from reviewing more than fifty of them. The things reviewers reject for, before they ever reach a reviewer.

  • Claims matched to evidence, line by line
  • The methodology holes reviewers find first
  • Figures, tables, and stats that survive scrutiny
  • Ethics, reproducibility, and venue requirements
Take the two minute readiness check

Download it free

A printable PDF. No email, no sign-up. Use it on your next paper and share it with your lab.

Download the checklist (PDF)

Learn the craft, end to end.

Hosted on Udemy. Leave your email on the courses page and you will hear the day it opens.

Opening soon

Research Methodology in AI: Experiments That Get Published

The full workflow from question to accepted paper. Experimental design, honest evaluation, and writing that gets past reviewers.

Ilhame Ait Lbachir

The real version of how research happens.

I am a computer science professor and AI researcher working on AI you can actually trust. Calibrated Researcher is where I share how that work really happens. Not the polished version. The real one.

Read the full story