Healthcare
17 columns
200 rows
8 formats

Medical Imaging Classification Dataset

This dataset provides comprehensive metadata for synthetic medical images, including diagnostic labels, imaging modalities, patient demographics, and workflow details. It is designed for deep learning applications in disease classification and radiology workflow automation, supporting robust model training and evaluation. The rich labeling and metadata structure enable advanced research in medical AI and workflow optimization.

last updated 20 Aug 2025
by GoMask

The brief that made it

Training and benchmarking deep learning models for disease classification

Sample rows

preview · 8 of 200 rows · all 17 columns

image_idstring	modalitystring	image_widthinteger	body_partstring	patient_idstring	acquisition_datedate	image_heightinteger	image_formatstring	diagnosis_labelstring	diagnosis_confidencefloat	radiologist_idstring	labeling_datedate	labeling_methodstring	disease_severitystring	ageinteger	genderstring	institutionstring
CT-2101-B1	CT	1024	abdomen	PAT-10001	2019-04-15	768	DICOM	appendicitis	0.88	RAD-201	2019-04-16	manual	moderate	16	female	Eastside Pediatric Imaging Center
MRI-3202-B1	MRI	640	brain	PAT-10002	2021-10-09	800	DICOM	stroke	0.74	RAD-308	2021-10-13	consensus	severe	72	male	Central Neurological Hospital
US-1123-B1	Ultrasound	1280	thyroid	PAT-10003	2023-02-21	800	PNG	nodule	0.63	RAD-215	2023-02-24	manual	mild	45	female	West Valley Endocrine Clinic
XR-4041-B1	X-ray	2048	chest	PAT-10004	2020-06-28	1024	DICOM	pneumonia	0.91	RAD-116	2020-06-29	manual	moderate	54	male	General Hospital of Chest Medicine
PET-5201-B1	PET	1024	whole body	PAT-10005	2022-01-07	2048	JPEG	metastasis	0.59	RAD-407	2022-01-09	consensus	severe	67	female	Oncology Imaging Institute
MG-6399-B1	Mammography	1200	breast	PAT-10006	2024-03-10	1600	JPEG	calcification	0.77	RAD-508	2024-03-11	manual	mild	48	female	North Regional Breast Care Research Center
CT-2102-B1	CT	1024	abdomen	PAT-10001	2019-04-15	768	DICOM	normal	0.96	RAD-201	2019-04-16	automated	none	16	female	Eastside Pediatric Imaging Center
US-1124-B1	Ultrasound	1280	thyroid	PAT-10003	2023-02-21	800	PNG	normal	0.92	blank	2023-02-24	automated	none	45	female	West Valley Endocrine Clinic
XR-4042-B1	X-ray	2048	chest	PAT-10007	2020-06-29	1024	PNG	tuberculosis	0.84	RAD-117	2020-07-01	manual	moderate	59	male	General Hospital of Chest Medicine
PET-5202-B1	PET	1024	heart	PAT-10005	2022-01-08	1024	JPEG	normal	0.97	blank	2022-01-10	automated	none	67	female	Oncology Imaging Institute
MG-6400-B1	Mammography	1200	breast	PAT-10006	2024-03-11	1600	DICOM	artifact	0.09	RAD-509	2024-03-12	manual	none	58	female	North Regional Breast Care Research Center
XR-4043-B1	X-ray	1024	hand	PAT-10008	2022-07-19	2048	PNG	fracture	0.54	RAD-118	2022-07-19	manual	severe	11	male	Children’s Orthopedic Imaging Clinic
MRI-3203-B1	MRI	512	spine	PAT-10009	2018-12-03	800	TIFF	herniated disc	0.68	RAD-309	2018-12-07	manual	moderate	63	male	Spine Health Imaging Center
XR-4044-B1	X-ray	1024	hand	PAT-10008	2022-07-20	2048	PNG	normal	0.93	blank	2022-07-20	automated	none	11	male	Children’s Orthopedic Imaging Clinic
CT-2103-B1	CT	2048	liver	PAT-10010	2020-01-15	2048	DICOM	fatty liver	0.66	RAD-202	2020-01-16	manual	mild	47	male	Metropolitan Digestive Disease Institute
CT-2104-B1	CT	2048	liver	PAT-10011	2020-01-16	2048	JPEG	normal	0.95	blank	2020-01-16	automated	none	47	male	Metropolitan Digestive Disease Institute
MG-6401-B1	Mammography	1600	breast	PAT-10012	2021-07-10	1200	JPEG	normal	0.99	blank	2021-07-11	automated	none	42	female	West Coast Women’s Imaging Center
MRI-3204-B1	MRI	512	pelvis	PAT-10013	2023-05-03	640	TIFF	osteoporosis	0.51	RAD-310	2023-05-07	manual	moderate	76	female	Senior Bone Health Imaging Center
US-1125-B1	Ultrasound	1024	liver	PAT-10014	2020-11-11	800	JPEG	artifact	0.06	blank	2020-11-16	automated	none	18	unknown	Edge Case Research Laboratory
XR-4045-B1	X-ray	2048	spine	PAT-10015	2019-09-30	1024	BMP	herniated disc	0.71	RAD-119	2019-10-03	manual	moderate	68	male	Spine Health Imaging Center

What the 200 rows show

from the 200-row sample

MRI modality stands out: mean image_width is 680.9, against 1,508 for the rest.

1,280median image_width
3labeling methods
4disease severities
4genders
6image formats
14diagnosis labels

Mean image_width by modality200 rows

image_width200 rows, in bands of 500

Median 1,280, from 512 to 3,072.

body_part200 rows · top 10 of 15 values

chest23
breast23
abdomen20
spine20
brain19
liver14
whole body12
pelvis11
kidney11
lung11

17 columns by typefrom the column list below

string 11
integer 3
float 1
date 2

Columns

17 columns in three groups

blueprint · 17 columns
column	type	description	example
Text 11 columns
`image_id`	string	Unique identifier for each medical imageunique	CT-2101-B1
`patient_id`	string	Unique identifier for the synthetic patient associated with the image	PAT-10001
`modality`	string	Imaging modality used (e.g., CT, MRI, X-ray, Ultrasound)7 values	CT
`body_part`	string	Anatomical region imaged (e.g., chest, brain, abdomen)	abdomen
`image_format`	string	File format of the image (e.g., DICOM, PNG, JPEG)6 values	DICOM
`diagnosis_label`	string	Primary disease or condition identified in the image (e.g., pneumonia, tumor, normal)	appendicitis
`radiologist_id`	string	Identifier for the synthetic radiologist who labeled the imageoptional	RAD-201
`labeling_method`	string	Method used for labeling (e.g., manual, automated, consensus)manual · automated · consensus · optional	manual
`disease_severity`	string	Severity of the diagnosed disease (e.g., mild, moderate, severe, none)none · mild · moderate · severe · optional	moderate
`gender`	string	Gender of the synthetic patientmale · female · other · unknown · optional	female
`institution`	string	Synthetic institution or hospital where the image was acquiredoptional	Eastern Renal Clinic
Numbers 4 columns
`image_width`	integer	Width of the image in pixels1 or more	1024
`image_height`	integer	Height of the image in pixels1 or more	768
`diagnosis_confidence`	float	Confidence score (0-1) for the diagnosis label0 to 1	0.88
`age`	integer	Age of the synthetic patient at the time of imaging0 to 120 · optional	16
Dates and times 2 columns
`acquisition_date`	date	Date the image was acquired	2019-04-15
`labeling_date`	date	Date the image was labeledoptional	2019-04-16

Use it for

A healthcare dashboard
Image_width by modality and a breakdown of body_part. Excel, Power BI or Tableau.
Why do the 36 MRI rows have a mean image_width of 680.9?
A root-cause class exercise
Hand out the rows and one question. The answer is in the data, not in the brief.
Images200
CT-2101-B11024CT
MRI-3202-B1640MRI
US-1123-B11280Ultrasou…
A software demo
Believable images with patient_id, modality and body_part to fill a screen in front of a buyer.

This dataset200 rows17 columns

Yours1,000 rows17 columns

blueprint · medical-imaging-classification-dataset

Behind this dataset

Same schema. As many rows as you need.

These 200 rows came out of a blueprint — 17 columns with generation rules behind each one. Open it in Data Factory to retune a column, add your own, wire in foreign keys, and run it at the size you actually need.

Rules it was built with

Include modality, image type, and diagnosis
Flag ambiguous and clear cases
Assign synthetic patient IDs
Link diagnostic confidence scores
Balance disease class distribution

Rows1,000 rows10,000 rows100,000 rows1,000,000 rows

Open the blueprint in Data Factory

1 credit per row. Preview 20 rows free. Buy rows for the full run.

Exports: CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence: yours to use, including commercially
API slug: medical-imaging-classification-dataset

What should your data show?

No signup. No card.

Sample rows

What the 200 rows show

Columns

Use it for

A healthcare dashboard

A root-cause class exercise

A software demo

Make it yours.

Same schema. As many rows as you need.

More in Healthcare

What should your data show?