This study demonstrates the scale of potential benefits of lossless image compression for a large district general hospital that does not currently implement compression in its PACS archive. We discuss the optimisation strategies that are used in cloud data storage to provide the context for baseline charges and image transmission performance that might be expected in modern cloud based image storage.

Analagously to increasing human obesity, our data stores are becoming heavier and hungrier by the day. Unlike the case for humans, this is happening largely invisibly in data-centres around the world. YouTube, TikTok, Instagram, WhatsApp and X – for example – are growing in content exponentially, and more rapidly than is medical imaging. All are unsustainable and there is widespread discussion about the growing use of electricity and water for data storage and processing, which includes AI.We are very used to dealing in gigabytes (GB)...

Scaled to an annual estimate for studies performed at our institution, the number of studies stored has increased over the period covered from 200 000 annually to more than 400 000, with a proportionally greater growth in cross-sectional imaging. This is in line with the national trend. The uncompressed average study size is close to 100 MB. With backups, an order of magnitude figure for current annual requirement in our hospital would be 100 TB. Our hospital performs approximately 1% of UK examinations giving a...

In this paper we have demonstrated the large gap between the current PACS storage in our hospital and what is possible, using cloud services to benchmark our needs. Compression is shown to offer significant storage and performance improvement potential. Automated archiving of rarely accessed studies (comprising in reality the majority of hospital imaging data) offers the potential of another order of magnitude reduction in expense that is not currently exploited.

D. Rosewarne: Nothing to disclose R. Marlow: Nothing to disclose M. Butt: Nothing to disclose q. wadood: Nothing to disclose F. Zaman: Nothing to disclose J. G. Zachariah: Nothing to disclose A. Jebril: Nothing to disclose H. Bilal: Nothing to disclose

The figure for the YouTube video daily upload volumes was from an LLM powered search and is regarded as credible by back-of-the-envelope calculation.