Imaging Modalities in Inflammatory Breast Cancer (IBC) Diagnosis: A Computer-Aided Diagnosis System Using Bilateral...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Inflammatory breast cancer (IBC) is an aggressive cancer with poorer survival rates.
• Current diagnostic criteria (AJCC) may not align with IBC presentation in North Africa.
• Limited expertise in IBC diagnosis due to its rarity poses challenges.

Purpose of the Study:

• To review current imaging modalities for IBC diagnosis.
• To propose a computer-aided diagnosis (CAD) system for early and improved IBC detection using bilateral mammograms.
• To address diagnostic discrepancies observed in North Africa.

Main Methods:

• A computer-aided diagnosis system was developed using bilateral mammograms.
• Type 1 and Type 2 fuzzy logic classifiers were employed.
• Key IBC markers analyzed include skin thickening, nipple retraction, and breast density asymmetry.

Main Results:

• The CAD system achieved high accuracy, ranging from 92.3% to 100%.
• Performance was evaluated using accuracy, recall, precision, F1 score, and AUC.
• The system focused on specific mammographic features relevant to IBC.

Conclusions:

• The proposed CAD system shows significant potential for early and accurate diagnosis of IBC.
• This approach can aid in overcoming diagnostic challenges, especially where IBC presentation differs from standard criteria.
• Further development could enhance diagnostic capabilities for this rare and aggressive cancer.