mednet.data.classify.indian

Indian database for TB detection (a.k.a. Dataset A/Dataset B).

The Indian collection database has been established to foster research in computer-aided diagnosis of pulmonary diseases with a special focus on pulmonary tuberculosis (TB). This database is also known as the “Database A/Database B” database.

  • Database reference: [noa14]

  • Split references: [noa14] with 20% of train set for the validation set

Important

Raw data organization

The Indian base datadir, which you should configure following the Setup instructions, must contain at least these two subdirectories:

  • DatasetA/ (directory containing the dataset A images in JPG format)

  • DatasetB/ (directory containing the dataset B images in DICOM format)

Data specifications:

  • Raw data input (on disk):

    • JPG RGB 8-bit depth images with “inverted” grayscale scale, with varying resolution of at least 1024 x 1024 pixels per sample

    • Samples: 156 images and associated labels

  • Output image: Use the same transforms and specifications as for classify.shenzhen

This module contains the base declaration of common data modules and raw-data loaders for this database. All configured splits inherit from this definition.

Module Attributes

DATABASE_SLUG

Pythonic name of this database.

CONFIGURATION_KEY_DATADIR

Key to search for in the configuration file for the root directory of this database.

Classes

DataModule(split_path)

Indian database for TB detection (a.k.a. Dataset A/Dataset B).

mednet.data.classify.indian.DATABASE_SLUG = 'indian'

Pythonic name of this database.

mednet.data.classify.indian.CONFIGURATION_KEY_DATADIR = 'datadir.indian'

Key to search for in the configuration file for the root directory of this database.

class mednet.data.classify.indian.DataModule(split_path)[source]

Bases: CachingDataModule

Indian database for TB detection (a.k.a. Dataset A/Dataset B).

Names of the JSON files containing the splits to load for montgomery and shenzhen databases (in this order).

Parameters:

split_path (Path | Traversable) – Path or traversable (resource) with the JSON split description to load.