Pydicom Medical Imaging

pip install pydicom numpy pillow

# Optional: compression codec handlers (install as needed)
pip install pylibjpeg pylibjpeg-libjpeg   # JPEG Baseline/Lossless
pip install pylibjpeg-openjpeg             # JPEG 2000
pip install python-gdcm                    # Comprehensive codec support

import pydicom
import numpy as np

# Read a DICOM file
ds = pydicom.dcmread("scan.dcm")

# Access metadata
print(f"Patient: {ds.PatientName}, Modality: {ds.Modality}")
print(f"Size: {ds.Rows}x{ds.Columns}, Bits: {ds.BitsAllocated}")

# Extract pixel data as NumPy array
pixels = ds.pixel_array
print(f"Pixel array shape: {pixels.shape}, dtype: {pixels.dtype}")

# Apply windowing for display (CT/MR)
from pydicom.pixel_data_handlers.util import apply_voi_lut
display = apply_voi_lut(pixels, ds)
print(f"Windowed range: [{display.min()}, {display.max()}]")

import pydicom

# Read DICOM file (defer_size delays loading large elements)
ds = pydicom.dcmread("scan.dcm")
ds_lazy = pydicom.dcmread("large.dcm", defer_size="1 KB")

# Access by attribute name (standard DICOM keywords)
print(f"Patient Name: {ds.PatientName}")
print(f"Study Date: {ds.StudyDate}")
print(f"Modality: {ds.Modality}")
print(f"Image Size: {ds.Rows} x {ds.Columns}")

# Access by tag number (group, element)
print(f"Patient ID: {ds[0x0010, 0x0020].value}")

# Safe access with getattr (avoids AttributeError)
slice_thick = getattr(ds, 'SliceThickness', 'N/A')
print(f"Slice Thickness: {slice_thick}")

# Iterate all elements
for elem in ds:
    if elem.VR != 'SQ':  # Skip sequences
        print(f"  {elem.tag} {elem.keyword}: {elem.value}")

# Read DICOM directory (DICOMDIR)
from pydicom.filereader import dcmread

dicomdir = pydicom.dcmread("DICOMDIR")
for record in dicomdir.DirectoryRecordSequence:
    if record.DirectoryRecordType == "IMAGE":
        ref_file = record.ReferencedFileID
        # ref_file is a list of path components
        print(f"Image file: {'/'.join(ref_file)}")

import pydicom
import numpy as np
from pydicom.pixel_data_handlers.util import apply_voi_lut, apply_modality_lut

ds = pydicom.dcmread("ct_scan.dcm")

# Basic pixel extraction
pixels = ds.pixel_array  # NumPy ndarray
print(f"Shape: {pixels.shape}, dtype: {pixels.dtype}")

# Apply Modality LUT (rescale to Hounsfield Units for CT)
hu_pixels = apply_modality_lut(pixels, ds)
print(f"HU range: [{hu_pixels.min()}, {hu_pixels.max()}]")

# Apply VOI LUT (windowing for display contrast)
display = apply_voi_lut(hu_pixels, ds)
print(f"Display range: [{display.min()}, {display.max()}]")

# Manual windowing (when VOI LUT metadata is absent)
center, width = 40, 400  # Soft tissue window
lower = center - width / 2
upper = center + width / 2
windowed = np.clip(hu_pixels, lower, upper)
print(f"Manual window [{lower}, {upper}]")

# Color images (ultrasound, photos) — handle YBR color space
import pydicom

ds = pydicom.dcmread("ultrasound.dcm")
pixels = ds.pixel_array
print(f"Color shape: {pixels.shape}")  # (rows, cols, 3)

# Convert YBR to RGB if needed
photo_interp = ds.PhotometricInterpretation
if "YBR" in photo_interp:
    from pydicom.pixel_data_handlers.util import convert_color_space
    rgb = convert_color_space(pixels, photo_interp, "RGB")
    print(f"Converted {photo_interp} -> RGB")

# Multi-frame (cine/video DICOM)
ds_multi = pydicom.dcmread("cine.dcm")
frames = ds_multi.pixel_array  # Shape: (num_frames, rows, cols)
print(f"Frames: {frames.shape[0]}, Frame size: {frames.shape[1:]}")

import pydicom
import numpy as np
from PIL import Image
from pydicom.pixel_data_handlers.util import apply_voi_lut

ds = pydicom.dcmread("scan.dcm")
pixels = ds.pixel_array

# Apply windowing
display = apply_voi_lut(pixels, ds)

# Normalize to 8-bit for standard image formats
if display.dtype != np.uint8:
    dmin, dmax = display.min(), display.max()
    if dmax > dmin:
        normalized = ((display - dmin) / (dmax - dmin) * 255).astype(np.uint8)
    else:
        normalized = np.zeros_like(display, dtype=np.uint8)
else:
    normalized = display

# Save as PNG
img = Image.fromarray(normalized)
img.save("output.png")
print(f"Saved output.png ({img.size[0]}x{img.size[1]})")

# Save as JPEG with quality control
img.save("output.jpg", quality=95)

# Batch conversion: directory of DICOM files to PNG
import pydicom
import numpy as np
from PIL import Image
from pathlib import Path
from pydicom.pixel_data_handlers.util import apply_voi_lut

def dicom_to_image(dcm_path, out_path, fmt="PNG"):
    """Convert a single DICOM file to standard image format."""
    ds = pydicom.dcmread(str(dcm_path))
    pixels = apply_voi_lut(ds.pixel_array, ds)
    dmin, dmax = float(pixels.min()), float(pixels.max())
    if dmax > dmin:
        norm = ((pixels - dmin) / (dmax - dmin) * 255).astype(np.uint8)
    else:
        norm = np.zeros_like(pixels, dtype=np.uint8)
    Image.fromarray(norm).save(str(out_path))

dcm_dir = Path("dicom_files/")
out_dir = Path("images/")
out_dir.mkdir(exist_ok=True)

for dcm_file in sorted(dcm_dir.glob("*.dcm")):
    out_file = out_dir / f"{dcm_file.stem}.png"
    dicom_to_image(dcm_file, out_file)
    print(f"Converted: {dcm_file.name} -> {out_file.name}")

import pydicom
from pydicom.uid import generate_uid

ds = pydicom.dcmread("original.dcm")

# Modify attributes
ds.PatientName = "Anonymous"
ds.PatientID = "ANON001"
ds.InstitutionName = "Research Lab"

# Add new attribute
ds.add_new(0x00081030, 'LO', 'Research Study')  # Study Description

# Delete attribute
if 'PatientBirthDate' in ds:
    del ds.PatientBirthDate

# Generate new UIDs for de-identification
ds.StudyInstanceUID = generate_uid()
ds.SeriesInstanceUID = generate_uid()
ds.SOPInstanceUID = generate_uid()

# Save modified file (preserves original)
ds.save_as("modified.dcm")
print(f"Saved modified.dcm with new UIDs")

# PHI anonymization: remove patient-identifying tags (DICOM PS3.15 Annex E)
import pydicom
from pydicom.uid import generate_uid

PHI_TAGS = [  # Core set — extend per institutional policy
    'PatientName', 'PatientID', 'PatientBirthDate', 'PatientSex',
    'PatientAge', 'PatientWeight', 'PatientAddress',
    'OtherPatientIDs', 'OtherPatientNames',
    'InstitutionName', 'InstitutionAddress',
    'ReferringPhysicianName', 'PerformingPhysicianName',
    'OperatorsName', 'StudyID', 'AccessionNumber',
]

def anonymize_dicom(ds, prefix="ANON"):
    """Remove PHI tags and assign anonymous identifiers."""
    for tag in PHI_TAGS:
        if hasattr(ds, tag): delattr(ds, tag)
    ds.PatientName, ds.PatientID = f"{prefix}_Patient", f"{prefix}_ID"
    ds.StudyInstanceUID = generate_uid()
    ds.SeriesInstanceUID = generate_uid()
    ds.SOPInstanceUID = generate_uid()
    return ds

ds = pydicom.dcmread("patient_scan.dcm")
anonymize_dicom(ds, prefix="STUDY001").save_as("anonymized.dcm")
print("Anonymized: PHI tags removed, UIDs replaced")

import pydicom, numpy as np, datetime
from pydicom.dataset import FileDataset, FileMetaDataset
from pydicom.uid import ExplicitVRLittleEndian, generate_uid

# File meta header
file_meta = FileMetaDataset()
file_meta.MediaStorageSOPClassUID = '1.2.840.10008.5.1.4.1.1.2'  # CT Image Storage
file_meta.MediaStorageSOPInstanceUID = generate_uid()
file_meta.TransferSyntaxUID = ExplicitVRLittleEndian

# Dataset with required attributes
ds = FileDataset("new.dcm", {}, file_meta=file_meta, preamble=b"\x00" * 128)
ds.SOPClassUID = file_meta.MediaStorageSOPClassUID
ds.SOPInstanceUID = file_meta.MediaStorageSOPInstanceUID
ds.StudyInstanceUID, ds.SeriesInstanceUID = generate_uid(), generate_uid()
ds.Modality, ds.Manufacturer = 'CT', 'Research'
ds.is_little_endian, ds.is_implicit_VR = True, False
dt = datetime.datetime.now()
ds.ContentDate, ds.ContentTime = dt.strftime('%Y%m%d'), dt.strftime('%H%M%S.%f')

# Pixel data from NumPy array
pixels = np.random.randint(0, 4096, (512, 512), dtype=np.uint16)
ds.Rows, ds.Columns = pixels.shape
ds.BitsAllocated, ds.BitsStored, ds.HighBit = 16, 12, 11
ds.PixelRepresentation = 0  # Unsigned
ds.SamplesPerPixel, ds.PhotometricInterpretation = 1, 'MONOCHROME2'
ds.PixelData = pixels.tobytes()

ds.save_as("new.dcm")
print(f"Created DICOM: {ds.Rows}x{ds.Columns}, {ds.BitsStored}-bit")

import pydicom
import numpy as np
from pathlib import Path

def load_dicom_series(series_dir):
    """Load and sort a DICOM series by slice position."""
    dcm_files = []
    for f in Path(series_dir).iterdir():
        try:
            ds = pydicom.dcmread(str(f))
            dcm_files.append(ds)
        except Exception:
            continue  # Skip non-DICOM files

    if not dcm_files:
        raise ValueError(f"No DICOM files found in {series_dir}")

    # Sort by ImagePositionPatient (z-coordinate) or InstanceNumber
    try:
        dcm_files.sort(key=lambda x: float(x.ImagePositionPatient[2]))
    except (AttributeError, IndexError):
        dcm_files.sort(key=lambda x: int(x.InstanceNumber))

    print(f"Loaded {len(dcm_files)} slices, "
          f"Series: {getattr(dcm_files[0], 'SeriesDescription', 'N/A')}")
    return dcm_files

def series_to_volume(dcm_files):
    """Stack sorted DICOM slices into a 3D NumPy array."""
    from pydicom.pixel_data_handlers.util import apply_modality_lut
    slices = []
    for ds in dcm_files:
        pixels = apply_modality_lut(ds.pixel_array, ds)
        slices.append(pixels)
    volume = np.stack(slices, axis=0)
    # Calculate voxel spacing
    pixel_spacing = dcm_files[0].PixelSpacing
    if len(dcm_files) > 1:
        try:
            z0 = float(dcm_files[0].ImagePositionPatient[2])
            z1 = float(dcm_files[1].ImagePositionPatient[2])
            slice_spacing = abs(z1 - z0)
        except (AttributeError, IndexError):
            slice_spacing = float(getattr(dcm_files[0], 'SliceThickness', 1.0))
    else:
        slice_spacing = float(getattr(dcm_files[0], 'SliceThickness', 1.0))
    spacing = (slice_spacing, float(pixel_spacing[0]), float(pixel_spacing[1]))
    print(f"Volume shape: {volume.shape}, Spacing (z,y,x): {spacing} mm")
    return volume, spacing

# Usage
dcm_files = load_dicom_series("ct_series/")
volume, spacing = series_to_volume(dcm_files)
print(f"HU range: [{volume.min()}, {volume.max()}]")

import pydicom
import pandas as pd
from pathlib import Path

FIELDS = ['PatientID', 'Modality', 'StudyDate', 'SeriesDescription',
          'StudyInstanceUID', 'SeriesInstanceUID', 'InstanceNumber',
          'Rows', 'Columns', 'SliceThickness', 'BitsStored']

def extract_metadata(dcm_path):
    """Extract key metadata from a DICOM file."""
    try:
        ds = pydicom.dcmread(str(dcm_path), stop_before_pixels=True)
    except Exception as e:
        return {"file": str(dcm_path), "error": str(e)}
    rec = {"file": str(dcm_path)}
    for f in FIELDS:
        rec[f] = str(getattr(ds, f, ''))
    return rec

# Scan directory recursively
dicom_dir = Path("dicom_archive/")
records = [extract_metadata(f) for f in sorted(dicom_dir.rglob("*")) if f.is_file()]

df = pd.DataFrame(records)
df.to_csv("dicom_manifest.csv", index=False)
print(f"Extracted metadata from {len(df)} files")
print(f"Modalities: {df['Modality'].value_counts().to_dict()}")
print(f"Unique patients: {df['PatientID'].nunique()}")

import pydicom, numpy as np, matplotlib.pyplot as plt
from pathlib import Path
from pydicom.pixel_data_handlers.util import apply_modality_lut

# Load and sort series by z-position
dcm_files = []
for f in sorted(Path("ct_series/").glob("*")):
    try: dcm_files.append(pydicom.dcmread(str(f)))
    except Exception: continue
dcm_files.sort(key=lambda x: float(x.ImagePositionPatient[2]))

# Stack into 3D volume (Hounsfield Units)
volume = np.stack([apply_modality_lut(ds.pixel_array, ds) for ds in dcm_files])
ps = dcm_files[0].PixelSpacing
z_sp = abs(float(dcm_files[1].ImagePositionPatient[2])
           - float(dcm_files[0].ImagePositionPatient[2]))
print(f"Volume: {volume.shape}, spacing: {z_sp:.2f}x{float(ps[0]):.2f}x{float(ps[1]):.2f} mm")

# Display orthogonal views (soft tissue window)
vmin, vmax = -160, 240  # center=40, width=400
fig, axes = plt.subplots(1, 3, figsize=(18, 6))
mid = [s // 2 for s in volume.shape]
axes[0].imshow(volume[mid[0]], cmap='gray', vmin=vmin, vmax=vmax)
axes[0].set_title(f"Axial (slice {mid[0]})")
axes[1].imshow(volume[:, mid[1], :], cmap='gray', vmin=vmin, vmax=vmax,
               aspect=z_sp/float(ps[1]))
axes[1].set_title(f"Coronal")
axes[2].imshow(volume[:, :, mid[2]], cmap='gray', vmin=vmin, vmax=vmax,
               aspect=z_sp/float(ps[0]))
axes[2].set_title(f"Sagittal")
for ax in axes: ax.axis('off')
plt.tight_layout()
plt.savefig("orthogonal_views.png", dpi=150, bbox_inches='tight')
print("Saved orthogonal_views.png")

import pydicom
from pydicom.uid import generate_uid
from pathlib import Path

PHI_TAGS = [
    'PatientName', 'PatientID', 'PatientBirthDate', 'PatientSex',
    'PatientAge', 'PatientWeight', 'PatientAddress',
    'OtherPatientIDs', 'OtherPatientNames',
    'InstitutionName', 'InstitutionAddress',
    'ReferringPhysicianName', 'PerformingPhysicianName',
    'OperatorsName', 'PhysiciansOfRecord', 'StudyID', 'AccessionNumber',
]

uid_map = {}  # Preserves study/series relationships across files
def get_mapped_uid(uid):
    if uid not in uid_map: uid_map[uid] = generate_uid()
    return uid_map[uid]

input_dir, output_dir = Path("original_dicoms/"), Path("anonymized_dicoms/")
output_dir.mkdir(exist_ok=True)
count, errors = 0, 0

for dcm_path in sorted(input_dir.rglob("*")):
    if not dcm_path.is_file(): continue
    try: ds = pydicom.dcmread(str(dcm_path))
    except Exception: errors += 1; continue

    for tag in PHI_TAGS:
        if hasattr(ds, tag): delattr(ds, tag)
    ds.PatientName, ds.PatientID = "ANONYMOUS", "ANON"
    ds.StudyInstanceUID = get_mapped_uid(ds.StudyInstanceUID)
    ds.SeriesInstanceUID = get_mapped_uid(ds.SeriesInstanceUID)
    ds.SOPInstanceUID = generate_uid()
    ds.save_as(str(output_dir / f"anon_{count:06d}.dcm"))
    count += 1

print(f"Anonymized {count} files, {errors} errors, {len(uid_map)} UIDs mapped")