Clinical benefits of BHI supported by 30 publications – new podcast

PubMed search now yields 30 publications on BoneXpert’s Bone Health Index, supporting the clinical benefits of BHI.

While the gold standard for assessing children’s bone health is DEXA BMD (Dual Energy X-ray Absorptiometry Bone Mineral Density), BHI leads to five complementary benefits:

  1. Properly Size-Corrected: The BHI formula was designed from first principles to be intrinsically independent of children’s highly variable bone size, overcoming a major confounding factor that limits the interpretability of DEXA’s areal bone mineral density.
  2. Z-scores by Bone Age: BHI is fully integrated with BoneXpert’s automated bone age determination from the same X-ray, allowing for the calculation of the Z-score (called BHI SDS, Standard Deviation Score) against bone age – a more accurate measure of physiological maturity than chronological age, in particular for children with growth and puberty disorders.
  3. Speed and Availability: As BHI is derived from routine bone age X-rays, it offers an opportunistic, rapid bone health assessment without additional radiation, cost, or patient visits, thereby increasing the frequency of monitoring in at-risk children.
  4. Clinical Validation: BHI SDS correlates with the widely used Height-Adjusted DEXA BMD Z-score (HAZ). And studies have demonstrated that a low BHI SDS is a significant predictor of fractures in high-risk populations, including children with Klinefelter syndrome, chronic inflammatory/disabling conditions and Duchenne muscular dystrophy (DMD).
  5. Unique Applicability: BHI is uniquely applicable in patient populations where DEXA is often impossible or unsafe. This includes children with severe physical disabilities (e.g., DMD, and cerebral palsy) due to positioning challenges and metallic implants, and children under three years of age, for whom BHI provides a sedation-free alternative to DEXA.

This podcast, generated by Notebook LM, unfolds these benefits

0:00
1:03

Study of BHI in children below 2 years with and without Osteogenesis Imperfecta

This study has a sinister background: In children below the age of two years with unexplained fractures, it may be difficult to distinguish those with low bone density due to conditions such as osteogenesis imperfecta (OI) from those who have been abused (by violence). Therefore, a new paper by Shalof and Offiah from University of Sheffield and Sheffield Children’s Hospital investigated whether bone health index standard deviation scores (BHI SDS) is sufficiently sensitive to distinguish between children with and without OI. The study included 33 radiographs of children with OI, and 89 of children with suspected abuse and not OI, i.e., normal children. The BHI SDS values of the normal children were compatible with a normal distribution with mean 0 and SD 1, showing that the BHI reference curves in BoneXpert, which were based on children born in 1955 in Paris, are still valid in a modern population. The OI children had a mean BHI SDS = –0.5 and SD around 1.

BHI SDS in boys with Duchenne muscular dystrophy (DMD) is associated with spinal fragility

A new paper by Drs. Phung and Ward from CHEO (Ottawa) reports a study of 54 boys with DMD with BoneXpert. Lateral spine radiographs were evaluated to derive the Spinal Deformity Index, a measure of fragility.

Lower BHI SDS was associated with higher spinal fragility. The authors concluded that BoneXpert provides a rapid method to identify those at greater risk of fracture without the need for additional imaging, and that BoneXpert can assist in identifying patients who may benefit from fracture prevention treatment.

BoneXpert validation study in Turkey

Researchers from Koc University Hospital in Istanbul have published a study validating two bone age systems: BoneXpert and Vuno.

292 images were rated with BoneXpert and Vuno Med bone age and compared to a reference formed by two manual ratings.

The accuracy of the two systems was found to be the same, as illustrated in the below Bland Altman plot for girls

Accuracy is one aspect of an automated method – the ability to explain the result is another, and here the two systems are very different, as the authors illustrated by juxtaposing the outputs from BoneXpert and Vuno:

BoneXpert’s main bone age result, 7.54 y, is derived as an average over the 21 tubular bones with equal weights on the bones. In addition, BoneXpert reports a carpal bone age – the average over the 7 carpals.

These results are “explained” by showing the contour of each bone as well as its bone age. Occasionally, a bone is left out due to abnormal shape, but not in this example.

Vuno’s “explanation” is a heat map, depicting where the deep neural net output is most sensitive to changes in the image. In this example, it showed a high intensity in radius, ulna, PP2 and DP3. Since the most reliable bone age is derived by averaging over as many bones as possible, it is of concern that the method collected information from mainly these four bones. BoneXpert’s output suggested that PP2 has bone age 9.0 y, much larger than the average 7.5 y. BoneXpert assigns the same weight to all 21 bones, which tends to average out differences between the bones and this provides precision and standardisation. In a follow-up exam, the Vuno method could – at its own discretion – decide to focus on a different small subset of bones, and this would raise the question whether a change in bone age was due a biological change, or merely a result of emphasising a different subset of bones.

The full article pdf is freely available here

Bone age validation study from Leipzig

Radiologists at the University Hospital in Leipzig have published a comparison of three automated bone age methods: BoneXpert, BoneView form Gleamer and PANDA from Image BiopsyLab

The study collected images of 306 children covering the age range 1–18 years. This wide range allowed the study to reveal how the methods behaved at low and high bone ages.

A reference bone age (denoted  “Ground Truth”) was formed as the average of three human ratings.

The overall agreement between each AI and the reference was almost the same for the three methods, although BoneXpert still showed the best correlation

BoneXpert is the only method that covers the full bone age range 0-19 yr, and it is interesting to see how the other methods handled the low- and high-bone age ends.

BoneView rejected images if the chronological age of the DICOM file was below 3 y and also rejected images with a bone age 17 and above.

PANDA accepted all images, but gave large errors below 5 years as is clearly visible. Also, for females with reference bone age above 15 y, PANDA tended to give predicted bone age not much larger than 15 y – a kind of saturation effect.

The corresponding author is Dr Daniel Gräfe and the full text is available online

AIFI pilot started in five Dutch hospitals

February 17, 2025 – Tiel – The AI For Imaging (AIFI) pilot project is officially launched. The AIFI project focuses on making artificial intelligence (AI) software for medical diagnostics accessible with the aim of reducing the workload for healthcare professionals and specialists via a shared national infrastructure.

The pilot involves a collaboration between five Dutch hospitals, the Dutch Association for Radiology (NVvR), and VZVZ, and is made financially possible by Healthcare Insurers Netherlands.

Continue reading

Automated bone age at University Hospital Crosshouse

Located in Crosshouse, near Kilmarnock in Scotland, University Hospital Crosshouse is a teaching hospital and the largest facility within NHS Ayrshire & Arran, serving a wide region. In 2023, the hospital adopted BoneXpert, an automated software solution for bone age analysis, to address workflow inefficiencies and improve diagnostic accuracy. As highlighted in the article by the Crosshouse Children’s Fund:

Continue reading

Google AI talks about BoneXpert AI

We might be crossing a line these days, as AI is not only taking over some of the radiologists’ work, but also taking over talking about it.

Listen to a podcast, which was generated autonomously by Google NotebookLM, from the pdf of the latest paper on the BoneXpert method for autonomous bone age determination.

  • The paper: Link
  • The podcast:
    0:00
    1:03

The podcast is surprisingly good at summarising the general aspects, but for an exact account of the accuracy, please consult the paper.